CIRP Annals - Manufacturing Technology 63 (2014) 773796

Contents lists available at ScienceDirect

CIRP Annals - Manufacturing Technology

jou rnal homep age : ht t p: // ees .e lse vi er . com /ci r p/ def a ult . asp

Design and management of manufacturing systems for production quality

Marcello Colledani a,b, Tullio Tolio (1)a,b,*, Anath Fischer (1)c, Benoit Iung (2)d,
Gisela Lanza (2)e, Robert Schmitt (2)f, Jozsef Vancza (1)g,h
a

Politecnico di Milano, Department of Mechanical Engineering, Via la Masa, 1, 20156 Milan, Italy
ITIA-CNR, Institute of Industrial Technologies and Automation, Via Bassini 15, 20133 Milan, Italy
CAD & LCE Laboratory, Faculty of Mechanical Engineering, Technion-Israel Institute of Technology, Haifa, Israel
d
Lorraine University/CRAN, Faculte des Sciences, Vandoeuvre, France
e
Institute of Production Science (wbk), Karlsruhe Institute of Technology (KIT), Kaiserstrasse 12, 76131 Karlsruhe, Germany
f
Laboratory for Machine Tools and Production Engineering WZL, RWTH Aachen University, Steinbachstr. 19, 52064 Aachen, Germany
g
Institute for Computer Science and Control, Hungarian Academy of Sciences, Budapest, Hungary
h
Department of Manufacturing Science and Technology, Budapest University of Technology and Economics, Hungary
b
c

A R T I C L E I N F O

A B S T R A C T

Keywords:
Manufacturing system
Production quality
Maintenance management

Manufacturing companies are continuously facing the challenge of operating their manufacturing
processes and systems in order to deliver the required production rates of high quality products, while
minimizing the use of resources. Production quality is proposed in this paper as a new paradigm aiming at
going beyond traditional six-sigma approaches. This new paradigm is extremely relevant in technology
intensive and emerging strategic manufacturing sectors, such as aeronautics, automotive, energy,
medical technology, micro-manufacturing, electronics and mechatronics. Traditional six-sigma
techniques show strong limitations in highly changeable production contexts, characterized by small
batch productions, customized, or even one-of-a-kind products, and in-line product inspections.
Innovative and integrated quality, production logistics and maintenance design, management and
control methods as well as advanced technological enablers have a key role to achieve the overall
production quality goal. This paper revises problems, methods and tools to support this paradigm and
highlights the main challenges and opportunities for manufacturing industries in this context.
2014 CIRP.

1. Introduction, motivation and objectives

Product quality and delivery reliability are key factors for
success in the manufacturing industry. Moreover, the increasing
emphasis on sustainable production requires maintaining the
resource efciency and effectiveness along the product, process
and production system life cycle [274]. Quality, production
planning and maintenance are fundamental functions for achieving these goals. They have been widely analysed in the literature
over the past several decades. The production planning eld has
developed methods for reducing work in progress (WIP), while
meeting desired production rate levels. The Statistical Quality
Control (SQC) eld has introduced optimized tools for monitoring
the behaviour of processes to achieve the desired product quality.
The Maintenance Management eld has derived policies for
preserving the efciency of degrading resources over time by
offering pro-active and predictive capabilities [112]. Traditionally,
all these elds have been treated by scientists and industrialists
almost in isolation. Yet it is clear that equipment availability,
product quality and system productivity are strongly interrelated.
As a matter of fact, quality, maintenance and production planning

* Corresponding author.
http://dx.doi.org/10.1016/j.cirp.2014.05.002
0007-8506/ 2014 CIRP.

strongly interact and jointly determine those aspects of a

companys success that are related to production quality, i.e. the
companys ability to timely deliver the desired quantities of
products that are conforming to the customer expectations, while
keeping resource utilization to a minimum level.
For example, low WIP improves the ability of identifying quality
problems in the system at an earlier stage but at the same time
makes maintenance actions more critical to the system. More
inspections make it possible to better assess the degradation state
of the resources yet also increase the system lead-time. Frequent
maintenance of resources tends to improve part quality, but
reduces the operational time of the machines in the system, which
affects the overall production.
It is clear, then, that the mutual relations among quality,
production planning and maintenance control should not be underestimated while conguring and managing the manufacturing system
as a whole. Important practical questions, such as Which is the
expected system effective production rate if the time to preventive
maintenance of one machine is reduced? and Which is the effect of
increasing the inspection frequency of one product feature on the
overall production yield of the system? remain unsolved. This lack of
understanding results in sub-performing unbalanced systemic solutions that tend to privilege one of the aspects while penalizing the
overall manufacturing system efciency.

774

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

The key questions that this paper addresses can be formulated

as follows: Which are the main industrial problems related to the
achievement of production quality targets? Which tools can
support the joint consideration of quality, production logistics and
resource maintenance in manufacturing system design and
operation? and Which are the main technical achievements
and insights brought by the use of these tools in industry?
Recently, several production paradigms have been proposed
that are strongly related to and have an impact on production
quality. These paradigms have considerably reshaped the boundaries within which the three aspects interact. Recongurability
[134], exibility [278], changeability [309] and co-evolution [280]
stress the importance of aiming at a strong coordination between
the dynamics of the system life cycle and the dynamics of the
product and process life cycles. Takata et al. [274] introduce the
notion of life cycle maintenance to be in phase with such
requirements. Yet, if a system evolves with faster and faster
dynamics, new challenges arise for production quality. In particular,
the long-term performance of the system becomes less important,
while production quality performance during system ramp-up
assumes fundamental relevance [86]. Moreover, small-lot production becomes more frequent than mass production. Therefore, a
new production quality paradigm is needed for mass customization
[60] and mass personalization [282], for open architecture
products [135] and for high product variety manufacturing [79].
Available concepts and programmes, such as Six-Sigma, Just In
Time, Continuous Improvement, Total Quality Management,
Toyota Production System and World Class Manufacturing, are
not designed for such dynamically changing contexts. A new
integrated concept of production quality needs to be developed to
meet this aim.
Another industrial trend that has been recently investigated and
framed [81] is the increase of the complexity of manufacturing
systems, in terms of hardware, software and management rules.
Complexity strongly undermines the achievement of the desired
production quality performance. Complex systems are typically
characterized by alternative process plans [208], multiple parallel
resources, part type dependent routings, and late variant differentiation [102]. The resulting challenge lies in the additional burden
placed on diagnosis, root-cause analysis, and error budgeting.
In response to these innovative aspects of manufacturing
systems, multiple in-line technologies for data gathering and
performance monitoring have emerged. A considerable amount of
data is typically made available on modern shop oors by multisensor technologies [304]. However, most of the time this
information is treated only locally and is not spread among
different company functions nor among partners within a
production network. For example, it is not infrequent for a quality
management department to ignore the reliability statistics of the
machines on the shop oor [152]. This behaviour makes it hard to
correlate disruptive phenomena taking place at shop oor level
with the product quality and to gather insights in the behaviour of
the system as a whole. It would be necessary to move from isolated
engineering practices to more integrated ones such as advocated
by System Engineering initiative [105]. Therefore, these data are
not fully exploited and translated into a business competitive
advantage for the company.
The impact of complexity on production quality is even more
signicant when considering the production network level. For
example, except for the period of the deep economic crisis 2009
2010, the number of recalls has been constantly increasing also due
the lack of inter-organizational quality systems [61]. Product recalls
indicate that manufacturing companies are particularly vulnerable
to ensure quality when they source via a global supply chain with
poor visibility [164]. Global automotive warranties are estimated at
USD 40 billion per year, i.e. a 35% loss in sales [89]. Low priced
production often leads to quality problems, and outsourcing leads to
a shift in knowledge concerning techniques and processes. Thus,
technical failures are more likely to occur due to communication
failures among the different parties engaged in the supply chain and

to missing denitions for technical interfaces. Since most of the aws

that eventually cause failures are introduced in the production phase,
early failure analysis can avert high recall costs and loss of image.
Legislation that limits industrial waste production, increases
target product recyclability rates and places the manufacturer at
the centre of the end-of-life treatment process through the
Extended Product Responsibility (EPR) principle is an additional
driver that strongly inuences the production quality paradigm by
penalizing the generation of defects and waste in manufacturing.
Moreover, sustainability issues related to energy efcient production [76] have to be taken into account while designing and
operating the system as a whole for a desired output production
quality-related performance target.
To promote intense and coordinated research activities aimed at
developing innovative technological and methodological solutions
to the aforementioned challenges, industrial organization and
funding bodies have recently launched several actions. For example,
at European level, the Factories of the Future (FoF) Public Private
Partnership has included the topic Zero Defect Manufacturing as a
priority in its FoF 2020 Roadmap. Moreover, under the FP7 call on
Zero Defect Manufacturing four projects have been funded
boosting cross-sectorial research and aiming at achieving the
largest possible target impact for the developed technologies. These
activities share the objective of supporting the development of a
knowledge-based manufacturing and quality control culture within
the EU. Similar activities have also been promoted in the USA within
the Advanced Manufacturing Partnership (AMP).
This paper provides an overview and a framework of the
industrial practices, scientic methodologies, and enabling technologies to protably manage the production quality targets in
advanced manufacturing industries. It also identies key open
research and practical issues that need to be addressed by the
research community. The paper is structured as follows: the next
paragraph presents a set of real cases that demonstrate the
industrial motivation to the problem. Section 2 proposes a new
system dynamics model for highlighting the relevant quality,
maintenance and production logistics interactions. Sections 3 and
4 discuss, respectively, the state-of-the art methods and tools and
the enabling technologies supporting the production quality
paradigm. Finally, Section 5 describes the most promising future
research topics in this area.
1.1. Industrial motivation
In order to highlight the main practical implications related to
the interactions among quality, production logistics and maintenance and to point out how these challenges are currently tackled
by companies, a comprehensive set of real industrial examples
have been collected. These case studies have been gathered by
analysing existing publications, running industrial projects, both
publically and privately funded, and by gathering authors
expertise. They include both traditional production sectors such
as the automotive and electronics sector and emerging sectors of
certain interest for the worldwide manufacturing context, including the production of medical devices as well as the green energy
production industry. Moreover, they include a reasonably wide
spectrum of manufacturing processes, such as machining, assembly and forming, at both macro and micro scales, and on both
metallic and non-metallic workpieces.
The industrial cases support the following considerations:
 The interaction among quality, production logistics and maintenance aspects is a complex issue to be managed.
 This problem involves different companies and different
departments within each company. The coordination and
cooperation among them in achieving a right balance between
these conicting goals is seen as a key issue for success.
 Depending on the specic product and market context, companies tend to prioritize one of the aspects. Finding the right
balance boosts the long-term company protability.

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

775

 The increasing complexity of products, processes, and systems is

one of the major challenges for production quality.
 Emerging ICT and inspection technologies as well as cooperation
based on risk-sharing contracts are seen as fundamental enablers
to meet production quality targets.
 Emerging production paradigms, such as recongurability and
exibility, pose new challenges for production quality.
 Industrial companies are experiencing a trend towards increased
investments in their ability to protably drive production quality
trade-offs.

1.2. Analysis of real cases

The rst example refers to the production of batteries for
electric vehicles in the electric mobility industry. The e-mobility
industry is expected to reach its target production by 2020.
As demand is still limited, manufacturers are putting great
emphasis on quality improvement. Dominating technologies to be
adopted for the production and inspection of batteries are still
lacking [143]. Error propagation is the major cause for defects.
Different quality tests take place such as electric test, stacking test,
leaking test, and temperature tests. However, not all the root
causes for defects are known since the quality management is still
in a learning phase. Therefore, a specic procedure is adopted to
manage the ramp-up during the introduction of new technologies
in the plant. The quality planning process starts with the
identication of critical product characteristics to be used to
determine the product quality level. First, new production and
inspection technologies are temporarily integrated off-line in the
factory to avoid interference with the cycle time of the main line
during the ramp-up phase (Fig. 1). In this phase, technology
improvement practices are implemented and knowledge of the
process behaviour is gathered. Once the process is made stable, the
technologies are developed as on-line integrated operations. In this
context (German BMBF Project ProBat [149]), the main relevant
questions are where to integrate the measurement, with which
technologies and which strategies? What are the consequences of
these choices on the quality and production logistics performance? Only by integrating quality management in factory
planning can these implications be captured.

Fig. 1. Procedure adopted by the company for managing the ramp-up of new
technologies in a plant assembling car batteries.

The needs for in-line inspection in multi-stage manufacturing

processes are also addressed in the following real-life example
related to the production of electrical engines for the automotive
industry at Robert Bosch GmbH (Fig. 2(a)). This real case is one of
the demonstrators in the MuProD FP7 EU-funded project [200]. The
proposed example is specically related to the assembly line of
electrical drives. This is a multi-stage system typically involving
2030 process stages. Three main ows are found, the rst
dedicated to the assembly of the magnetic rotor, the second related
to the assembly of the stator and the last related to the coupling of
stator and rotor to produce the complete engine.

Fig. 2. Electric drive produced at Robert Bosch GmbH (a) and Planet Carrier
produced at Gamesa (b).

State-of-the-art inspection technologies facilitate assessing the

quality of the engine by end-of-line testing of several product
features, the most important being the magnetic torque of the
rotor. However, in order to better understand the causes for
deviations and to allow process control and improvement at early
process stages, innovative inspection technologies need to be
developed and distributed in the upstream rotor assembly process
stages. The rotor is composed of a set of magnets assembled on the
surface of multiple laminated stacks. These stacks are then axially
assembled to produce the rotor. The number of assembled stacks
determines the specic product type.
Knowing the effect of stack magnetization on the rotor
magnetic intensity and, ultimately, on the nal engine torque is
a major issue in this manufacturing process. This would allow
characterizing the correlation between production stages, consequently controlling the upstream stack assembly strategy to obtain
the desired engine quality levels. Another challenge is to determine
ad-hoc assembly strategies that can prevent the propagation of
defects from the early stages to the nal assembly stage. In
MuProD, one of the considered options exploits the quality
correlation between the stage where the stacks are magnetized
and the stage where the rotor is assembled. A defective stack can be
turned into a good quality rotor if the assembly angle is suitably
compensated at the downstream stage.
The second considered solution is to integrate selective and
adaptive assembly strategies in the rotor assembly system [126].
Selective assembly entails on-line part inspection, clustering parts
into bins according to specic key quality characteristic values and
subsequent matching only from coupled classes according to some
predetermined matching criterion. This approach makes possible
to change a product quality problem into a system design and
operation problem. In the case of rotor assembly the introduction
of selective assembly can increase production quality signicantly
by reducing scrap and incrementing the yield of the system.
The third example refers to the manufacturing of small-lot large
parts (i.e. planet carriers) for windmill gear boxes in the wind
power sector at Gamesa (Fig. 2(b)) [200]. The continuously
increasing demand for energy is leading to the manufacturing of
eolic towers that are able to produce more power. These towers
demand larger components and require new and lighter materials
for easier assembly. The machining of components such as the
planet carrier is critical, since very small product features have to
be machined at very tight tolerance requirements [93] (normally
tenth of microns on dimensional and geometrical features) on very
large parts (outer diameters up to 2500 mm, weight up to 7000 kg).
The production system adopted in the reference case is composed
of parallel machining centres dedicated to roughing and nishing
operations. The causes of defects are related to the input casted
material, part deformation due to xturing, tool wear, vibrations,
etc. In order to achieve such highly demanding manufacturing
goals, the company makes use of a hybrid inspection procedure.
The rst part of the lot is extensively measured at the CMM for
compensating possible deteriorations by machining parameter
adjustment. Then, the lot production is started. For each processed
feature, extensive in-process part verication is carried out to

776

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

avoid the generation of any possible type of defect, due to the high
value of casting parts. However, these continuous machining,
measuring and adjustment loops interfere with the cycle time and
the productivity of the plant. Therefore, this is an example where
the solution adopted by the company is strongly polarized on
quality performance, with negative consequences for production
logistics performance. This approach is also evident in other
sectors, such as the production of critical mechanical components,
i.e. engines, in the aeronautic industry.
A fourth example is related to the production of customized
micro-intravascular catheters as high value medical products for
the ageing society in the medical technology sector, as at ENKI S.r.l
in Italy (Fig. 3).

Fig. 3. Multi-lumen and multi-layer catheters for medical applications.

Medical technology is one of the most innovative industry in

the world, with an increase of 15% in turnover since 2005. As
opposed to the previous case, this example shows a completely
different manufacturing context that is related to micro-machining and micro-forming operations and one-of-a-kind customized
products. These types of catheters have applications in oncology,
angiology, angiography and angioplasty, where the demand for
customized single-use products is increasing to solve sterilization
problems and to reduce the risk of contamination inherent in
multiple-use products. Over the last years, a trend towards
miniaturization of these devices is in place. The successful
achievement of this goal will facilitate the use of these catheters
in smaller arteries, thus having a great impact on the number of
curable diseases and ultimately saving lives.
These micro-catheters are composed mainly of a micro-tube
and, depending on the specic application, an injection moulded
part that makes it possible to carry out the surgery. The microtubes can be either single lumen or multi-lumen facilitating the
transportation of multiple substances to the zone of interest, as
well as single layer or multi-layer for high-pressure resistance. The
manufacturing process is composed of four main phases: (i)
material compound preparation and control, (ii) micro-machining
of the extrusion die (micro-milling, micro-Electrical Discharge
Machining EDM) for each specic part type, (iii) micro-extrusion
of the micro-tubes and (iv) nal micro-catheter assembly. The
major causes of defects are related to defects in die production that
cause defects in the micro-tubes and geometrical defects
generated within the micro-extrusion process.
The above defects lead to an extremely high defect rates (up to
70% in standard production). These defects are certied mainly by
100% micro-tube inspection at the end of the line, which is
manually driven and expensive. This high defect rate also
undermines the possibility of robust production scheduling and
is translated directly into service level issues. Moreover, this huge
amount of generated scrap results in a massive waste ow, which
is an additional cost for the company that must pay for its
treatment. This example shows how in the context of high process
variability, poor controllability and automatic inspection, as well
as relatively low material value, the company strategy may be

highly polarized on productivity performance, thus penalizing

process control and rst-time-right quality strategies.
Another example is related to the recently designed engine
block production line at Scania CV AB, Sweden. Scania is a
worldwide manufacturer of trucks and buses. All Swedish
production was recently moved to Sodertalje. A serial production
line layout with multiple parallel machining processes per stage
has been designed with the goal of producing different engine
block types in the same system at a very high production rate. As a
result, some 400 product paths are possible while considering all
possible routing alternatives in the system. The adoption of parallel
processes increases the reliability of the system, thus making it
possible to reach increased productivity targets. Nevertheless, this
poses additional challenges regarding quality control and part
deviation verication with respect to serial system layouts. Indeed,
multiple product paths generate a mixing effect, loss of process
signature and loss of FIFO rules, thus reducing traceability in the
system, i.e. the ability to connect a defect with the process that
generated it. Moreover, in the presence of end-of-line inspections,
long delays in quality feedback are generated. This clearly reduces
the ability to close a reactive quality control loop but increases the
total production rate of the system. Therefore, in order to increase
the visibility of quality and process deviations, in-line inspection
points need to be distributed that will have a positive impact on
quality and a negative impact on production logistics performance.
This example proves that manufacturing system design affects
product quality and that product inspection design affects the
production logistics performance of the system.
The assessment of customer perception of products in multistage manufacturing systems is one of the main challenges of
production and quality engineering and the main topic of the BMW
Group case study [244,248]. This real case study is also
demonstration scenario of the Cluster of Excellence Integrative
Production Technology for High-Wage Countries [240]. The vehicle
acoustics is a product feature that is important to the customer
perception of the product quality. It has very complex and
multifaceted mechanisms that generate structure-borne sound,
which is then transferred to the interior of the vehicle via the car
body. When the noise reaches a particular level inside the car, it
may be perceived by the customer as annoying. The technical
analysis shows that the rear axle drive has a pronounced effect on
the acoustics within the vehicle (see Fig. 4).

Fig. 4. List of requirements on the hypoid gear sets for passenger vehicle axle drives
of standard design.

In order to ensure this customer requirement, advanced tools

for inspection planning and quality control methods in multi-stage
production systems are required. The manufacturing of rear axle
drives, is characterized by many variants, which are produced at

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

medium lot sizes (10,000 units). Different manufacturing

processes are used for different variants. The frequent change in
variants leads to a high planning effort for the necessary adaptation
of the manufacturing processes. The analysed process chain is
shown in Fig. 5. The main challenges in this multi-stage
production line are:

Fig. 5. The BMW production process, characterized by many sensitive tolerances

with complex dependencies.

 measurement of the acoustic perception of the customer;

 parameterization of the customer specication for in-line
process inspection;
 challenging assembly concept from FIFO production towards a
tolerance optimized assembly concept.
In order to close the loop between customer perception,
inspection planning and quality control, rst of all perceived
quality methods for the measurement of customer perception
regarding the acoustics of the rear axle are needed. By identifying
the relevant process parameters in a dynamic in-line inspection
the acoustic behaviour can be then forecasted. Even when inlineinspections are installed and all processes are stable deviations in
the multi-stage production system can cause critical acoustic
behaviour after the differential assembly. Hence, advanced
tolerance-oriented part matching tools could forecast the
acoustic t of the gear wheel and ring gear pair, assembled to
the gear set. This means both expanding any unnecessarily tight
tolerances to save costs and dening the critical tolerances more
precisely to ensure the desired functionality of the end product.
In order to reduce scrap rates the concept of just-in-sequence
production has to be extended to a tolerance-oriented production control principle, where production and quality control are
integrated.
The production quality paradigm in contexts characterized by
deteriorating products, such as fresh food or yoghurt, is addressed
next. Food production is pervaded by strict requirements on
hygiene and delivery precision. The production plants have to
quickly respond to the market demands and current order
situation. A typical production sequence for yoghurt includes
mixing/standardizing of milk, pasteurization, fermentation, cooling, addition of fruit additives and packaging. The production
planning involves very complex problems due to the maximum
allowed storage time before packaging. If the time the product
ows in the system exceeds this limit, it has to be scrapped.
Changeovers are typically sequence-dependent (increasing fat
level is preferred giving shorter set-ups) and up to 25 products
variants may be produced in the same system, with different
processing times. In these plants the primary objective is to control
the production of the different products for reducing the
changeover number (typically 100/week) and time (typically
20 h/week). Secondly, the objective is the reduction of the product
scraps (typically around 10%) due to obsolescence of inventory by
achieving a better synchronization of the process phases, an
effective joint control of the tank sizes (buffers) and the product
quality. The removal of this bottlenecks and the reduction of WIP is
a priority for these industries. Therefore, inventory management
and line balancing play a fundamental role in achieving the
production quality targets.

777

In automotive paint shops [120], to ensure high paint quality,

multiple inspection stations are usually allocated along the
cleaning and painting processes. Vehicles failed in inspection will
be repaired or repainted before moving to the next station.
Therefore, to improve the performance of paint shops, reducing
quality failure rates while keeping the production rate within the
target is of signicant importance. An automotive paint shop
typically consists of the following major processes. In the pretreatment section, each vehicle body is submerged in a phosphate
liquid to get a layer of coat on the surface of the steel. In the next
ELPO process (electro coat paint operation) the body of the vehicle
is covered with a special substance to protect it from corrosion.
Then, the body needs to be heated and baked in the ELPO oven, and
nally sanded to nish the ELPO process. Afterwards, there is an
intermediate stage where the pre-treatment quality is inspected.
The vehicle is then moved to the sand section followed by seal
inspection. The next is the painting section that starts by spraying
primer on the vehicle, which improves the adherence of the paint
to the vehicle body. Afterwards, base coat and clear coat are
performed. Then, the body of the vehicle needs to be baked. After
this process, the nal inspection (nesse) and, in case a defect is
detected, the repair processes are applied. Here, defects, such as
scratch, dirt, dent will be identied and xed. After repair, these
vehicles are sent to the next operations. In automotive paint shops,
imperfect dirt cleaning in the upstream sanding operations will
result in more paint defects in downstream colour coatings.
Therefore, the stage correlation and the management of defects
through part re-processing are the main issues to be addressed at
system level.
Production quality is of signicant importance also in the
semiconductor industry and, specically, in wafer fabrication. A
semiconductor manufacturing process has the following characteristics. The production is performed through multiple stages.
Some of these stages work in batches, including the slicing process,
lapping, and polishing. Multiple parallel processors are commonly
adopted to achieve the required production rate. Each product may
undergo several re-entry loops in the system. The production yield
is generally very low (around 50%) and the requirements on duedate performance are very strict. The ow time is extremely high
thus mining the reactiveness of the quality control system. High
priority lots typically share the production resources with low
priority lots, thus generating non-FIFO production sequences. In
this context, the complexity is the major barrier for production
quality.
1.3. The production quality paradigm
In the literature as well as in the industrial practice there are
many different Key Performance Indicators (KPIs), or performance
measures, that individually relate to quality, production logistics
and maintenance. In the following, the most widely adopted KPIs
at system level are considered. In manufacturing systems they are
complex non-linear functions of single process or single stage KPIs.
Typical system level production logistics KPIs include:
 The production rate, i.e. the number of parts produced in a given
time (also called throughput). It is usually measured in terms of
Jobs Per Hour (JPH).
 The total inventory, or WIP, i.e. the total amount of parts owing
in a system.
 The ow time, i.e. the time required for parts to cross the system.
 The interdeparture time, i.e. the time between consecutive
deliveries of output products.
These performance measures can be considered in the long
term or in the short term. Moreover, the rst moment (mean) or
higher moments of these measures can be taken into account. The
consideration of higher moments in the short term can be used, for
instance, to evaluate the so-called due-date performance. For
example, the service level of a system, which is the probability of

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

778

delivering a lot of a certain size X before its xed deadline T, is a

due-date performance. From a quality-oriented point of view,
typical KPIs of interest are:
 The system yield, or quality buy rate, i.e. the number of
conforming parts delivered by the system over the number of
conforming parts going into that system, in a specied period
of time. In case of 100% conforming input ow, it is simply the
fraction of good parts delivered by a system.
 The rst-time quality, or rst-time right rate, or rst-pass yield,
i.e. the good job ratio of all the rst-time processed jobs.
 The defect rate, i.e. the fraction of non-conforming jobs delivered
by the system.
From a maintenance point of view, typical system KPIs include:
 The system availability, i.e. the time a system is capable of being
operational in a given total time.
This analysis shows a fundamental lack of a clear taxonomy for
integrated quality, production logistics and maintenance performance measures [122,141]. An attempt towards the formalization
of a taxonomy has been recently proposed in [239]. Moreover, the
Total Quality Management (TQM) and Total Productive Maintenance (TPM) paradigms have proposed integrated KPIs to evaluate
the effectiveness of the implementation of a specic improvement
plan in industrial contexts. Although TQM and TPM share a lot of
similarities, are in fact considered as two different approaches in
the literature. TQM attempts to increase the quality of goods,
services and concomitant customer satisfaction by raising
awareness of quality concerns across the organization. Total
Productive Maintenance (TPM) is a system of maintaining and
improving the integrity of production machines that add business
value to the organization. These methodologies suggest that the
most relevant integrated performance measure is:
 the effective throughput, or the net throughput, also called OEE
(Overall Equipment Effectiveness), that is the number of
conforming parts produced by the system in a given time.
Grounding on this background knowledge, the production
quality paradigm can be formulated in the following terms:
Production quality is the discipline that combines quality,
production logistics, and maintenance methods and tools to
maintain the throughput and the service level of conforming parts
under control and to improve them over time, with minimal waste
of resources and materials.
2. Quality, production, and maintenance Interaction Model
Several empirical studies have discussed the interaction among
quality, production logistics and maintenance in manufacturing
systems. For example, in [27] a survey approach is used to identify
potential correlations between the application of JIT and TQM lean
practices in the automotive and electronic industries. The main
result of this analysis is that those companies that are more
successful in limiting their inventory and in better organizing their
production through JIT policies also achieve better quality
performance and apply more effective defect reduction programs.
This positive correlation highlights the need for a deeper
understanding of the interaction dynamics between these relevant
aspects in manufacturing.

this interaction. Among these, a very powerful set of tools is

business and system dynamics Causal Loop Diagrams (CLD). These
tools have been proposed for modelling complex interactions
between quantitative and qualitative variables in a number of
complex business management problems. Their application to the
analysis of the interactions between quality, maintenance and
productivity performance indicators is reported in [117,230,270].
The main goal of these models is to identify all possible
interactions among variables and decisions in order to support
the denition and implementation of continuous improvement
programs that do not fail to meet the goals due to unexpected
interactions. Understanding the relevant interactions then makes
it possible to avoid local improvements that deteriorate the global
performance due to neglected impacts. CLD charts are diagrams in
which the relevant variables of a problem are listed and connected
by directed arrows. In this format, A pointing at B with a positive
arrow means that, given that everything else is xed, an increase of
variable A causes B to increase more than it would normally. A
pointing at B with a negative arrow means that, given that
everything else is xed, an increase of variable A causes B to
decrease more than it would normally. Only direct and easily
explicable cause-effect connections have to be reported. CLD are
very powerful tools for nding existing control loops in complex,
multidisciplinary and dynamic contexts and in making them
explicit.
Although they have been widely used for consulting activities
and for policy making, state-of-the-art CLD models do not focus on
the production system design and operational levels but rather try to
see the problem more generally from a managerial point of view. For
example, the implications of continuous improvement programs
that take into consideration worker motivation, learning cycles and
company emphasis achieving performance targets, have been
investigated. One of the highlighted loops is the following: More
defects reduce net process throughput (effective throughput in our
notation). This in turn increases the actual versus target throughput
gap. This negative performance increases worker effort, which
positively affects gross process throughput (total throughput). This
in turn has a positive impact on net process throughput. This
reinforcement loop is called the Work Harder loop.
Although this dynamics plays a relevant role in the achievement of satisfactory production quality performance, the goal of this
keynote paper is to consider manufacturing and shop oor related
aspects. Therefore, a new interaction model is needed with the
specic goal of answering the following question: What are the
cause-effect relations explaining the mutual interactions among
quality, maintenance and production logistics in manufacturing
systems? Based on the real-life examples provided in the previous
section, in this paper an interaction model is developed and
proposed. The main objective of the proposed model is to dene
and characterize all major sources of interactions affecting
production quality at the shop oor level. These interactions are
consistent with the Functional Enterprise-Control Model as
proposed by the IEC/ISO 62264 standard [111].
The aggregated representation of the model is reported in Fig. 6.
This simple graph shows that bi-directional mutual cause-effect
relations can be found among quality, maintenance and production
logistics. A more comprehensive denition of these links in
manufacturing systems is provided in the detailed CLD model

2.1. The Interaction Model

The complex dynamics of the interactions among quality,
production logistics and maintenance requires considerable effort
to be modelled and understood. This activity is important to
identify and explain the many existing trade-offs. The literature
includes models developed to capture and explain the dynamics of

Fig. 6. High-level representation of the Interaction Model.

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

depicted in Fig. 7. The red, blue and green regions refer to variables
related to production logistics, quality control and maintenance,
respectively. The links of greatest interest in this paper are
represented by those arrows that cross regions of different colours.

779

leads to high inspection time and, consequently, low total

throughput. With reference to the ENKI case, the wide mix of
personalized parts and the small lot sizes lead to extensive set-ups.
Set-ups are detrimental for the total throughput and for process
deviations, thus leading to consistent generation of defects. In
addition, the poor process data gathering undermines the
possibility to observe the equipment degradation state, thus
making preventive maintenance hard to be implemented. This, in
turn, leads to short planned equipment downtimes, thus high total
throughput, but also to consistent process deviations, thus high
defect generation and low yield.
As these few examples show, the proposed Interaction Model
can be used by scientists to identify relevant unexplored problems
that need to be further investigated, as well as by practitioners to
motivate and gather insights on unexplained phenomena on the
boundaries of these three areas. In this paper, this model will be
used as a reference framework to structurally explore the topics
already addressed in the literature and to highlight promising
research areas for the future.
2.2. Using the Interaction Model to classify the scientic literature

Fig. 7. Details of the Interaction Model developed in this paper.

For example, following the arrows in Fig. 7, increasing the WIP,

in turn increases both the total throughput and the lead-time of the
system. The increase in lead-time, however, also causes qualityrelated phenomena. Indeed, it reduces the observability of the
critical product quality feature in the system, which in turn
reduces the ability to detect a potentially generated defect within a
short time. This directly translates into the propagation of more
defects between the processing stages in the system and a waste of
production capacity in processing parts that are already defective.
A loss of capacity leads to a loss of total throughput. This example
of balancing loop highlights the importance of this approach. If the
effects of the WIP increase on the product quality were overlooked,
quality and production logistics would be treated in isolation and
erroneous design and management decisions could easily be made.
Focusing on maintenance and production logistics interactions,
a second link made explicit by the diagram in Fig. 7 is explained in
the following. Equipment condition-based preventive maintenance is typically supported by sensorial data collected from the
eld while the equipment is operational. If these data are properly
analysed, they can be used to make inferences about the
degradation state of the equipment. If the sampling frequency
of this monitoring system is increased, the ability to detect the
equipment degradation state increases. This increases the chances
that an undesired degradation state will be correctly identied and
preventive maintenance practices will be promptly activated, thus
increasing equipment reliability, decreasing the frequency of
unexpected random failures and ultimately decreasing corrective
maintenance interventions. This maintenance-related loop also
has an impact on system logistics. Indeed, less corrective
maintenance generates less unplanned equipment downtime,
while more preventive maintenance causes an increase in planned
equipment maintenance interventions. Theses both affect the
production rate of the system. If this interaction is overlooked,
overall myopic decisions can be taken.
In the following, the relevant phenomena characterizing two of
the real cases investigated in Section 1.2 are framed within the
Interaction model. These few examples show that real cases can
be mapped within this Interaction Model. With reference to the
Gamesa case, the high workpiece inspection effort leads to high
observability of the product quality characteristics. This positively
affects the ability to detect and correct defects as soon as they are
generated. This is benecial for the system yield but detrimental
for the total throughput, as the production resources are used to reprocess parts and correct defects. The high inspection effort also

About 300 papers, mostly from leading international journals,

have been classied and framed within the Interaction Model
(Fig. 8). Specically, the papers have been clustered according to
two-dimensional axes. The rst axis relates to the specic phase
in the paper where the interaction is addressed. More specically,
the design and planning phase and the operational, control and the
management phase have been taken into account. The second axis
relates to the type of interaction addressed. According to the
proposed interaction model, possible interactions are qualityproduction logistics interaction, production logistics-maintenance
interaction, quality-maintenance interactions, and complete interaction among quality, production logistics and maintenance. In
Fig. 8, the bullets represent the cluster of papers addressing
common problems as framed within the interaction model. The
size of the bullet represents the population magnitude of the cluster.
Most of the contributions cover areas related to the interaction
among quality and production logistics, while only a few
contributions address problems under a fully integrated view.

Fig. 8. Paper distributions within the Interaction Model.

3. Problems and methodologies

3.1. Design and planning phase
In the following sections, the existing literature addressing the
links between quality, productivity and maintenance in the design
phase are revised. The focus is mainly on system and process
related design decisions while product and tolerance design
decisions, in spite of being important factors within the production
quality target, are not explicitly considered in order to avoid
deviations from the main scope on this paper. For a recent review
on the link between product design and quality see [180].

780

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

3.1.1. Impact of manufacturing system design on quality

There are many aspects that demonstrate that the production
system architecture affects the production quality performance, as
highlighted in [109] based on the analysis carried out in General
Motors. This review has been updated in [107], where recent works
focusing on this specic link have been framed.
In mass production systems the simulation works proposed in
[25,165] contributed to the assessment of a fundamental principle:
while changing the system architecture the fraction of conforming
products may drastically drop. A similar result was achieved by
[136]. The authors compared six alternative congurations,
including serial and parallel lines, and hybrid congurations, in
terms of multiple performance measures, including the system
capability to produce parts with limited variations and the
expected availability. The authors show that serial lines perform
better than parallel lines in terms of dimensional variation of
products, because there is only one possible path in the part ow
and the mixing effect is avoided. The mixing effect means that
multiple processing stages show different degradation patterns
and actual capabilities and this phenomenon increases the
variability in the key quality characteristics of the output products.
The mixing effect in parallel machine lines has been further studied
in [233]. The authors analysed by simulation the consequence of
the mixing effect on the ability of performing a root cause analysis
at the inspection points in the system. Other undesired phenomena, such as possible job order loss and sampling frequency
mismatch, have also been identied in parallel processes. In
[38,94,189] it was shown that U-shaped lines may perform better
than serial lines in terms of quality of the released output. The
reason is that the operators assisting the line can visually detect
quality problems in the system earlier than in serial lines and,
consequently, can react more promptly to these defects.
The impact of buffers on production quality has also been
analysed in the literature. The Lean Production area has shown that
the reduction of inventory has a positive impact on product
quality, since quality defects are identied earlier and are not
propagated throughout the system stages [310]. As a matter of fact,
Toyota Production System (TPS) advocates see in-process stocks as
waste (muda), which often hides production problems. However,
from the Manufacturing System Engineering area it is known that
the production rate of the system is positively affected by the
presence of buffers, since they decouple the behaviour of the
unreliable machines [65]. This trade-off has been studied
analytically in [52] and [131] from an integrated quality-logistics
point of view. The authors found cases in which the effective
throughput is maximized for a given buffer capacity. This
behaviour is due to the coupling of two contrasting effects, in
the presence of remote or ubiquitous inspections, where a product
feature manufactured at a certain processing stage is inspected at
a monitoring station located further downstream in the line. One
effect is the positive impact of the buffer capacity on the total
throughput of the system. The other effect is due to the delay of the
quality information feedback when remote inspection is performed. Processed parts do not instantaneously reach the
inspection point, but are stored in the inventory queue before
being measured. Large buffers between the monitored station and
the inspection point increase the time parts spend in this portion
of system. This causes long reaction time in identifying out of
control conditions and decreases the system yield. This behaviour
generates interesting considerations on the joint design of buffers
and quality control parameters in manufacturing lines [51]. Nada
et al. [201] developed a comprehensive framework to address the
aforementioned issue during the design phase of manufacturing
systems. A Congurator Capability Indicator (CCI) is developed, by
using hierarchical fuzzy inference, to select the most proper
architectural parameters of the system, under production quality
considerations.
The design of in-process buffers has a relevant impact on the
product quality also in those industries producing perishable or
deteriorating components and products. The quality characteristics

of perishable products deteriorate over time. For example, as

commented in Section 1.2, in food industry there is a maximum
storage time before packaging. The product has to be scrapped if
the time spent in the system overpasses a certain xed limit. This
problem has been addressed in [169]. A project to determine costefcient ways of speeding up the croissant processing lines of
Chipita International Inc. is reported. The installation of a properly
sized in-process buffer led to a reduction in failure impact on
product quality and an increase of the system efciency. In [168]
the authors focused on the production rate of asynchronous
production lines in which machines are subject to failures. If the
failure of a machine is long enough, the material under processing
in the upstream machines must be scrapped by the system. In
[295] a transient analysis is proposed to design the size of the
buffers needed in dairy lling and packaging lines. The distribution
of the ow time in unreliable multi-stage manufacturing systems
was evaluated in [260]. This method can support the design of
buffers for achieving a certain accepted scrap rate in perishable
good production. In [272] an inventory model for perishable
products with random perishability and alternating production
rate is proposed. As shown in these works, buffers should be
designed by using an integrated production quality oriented
approach.
In machining and assembly operations it has been shown that
the design decisions concerning the system operating speed are
strongly correlated to the product quality [214]. Improving the
machines processing rate has a positive impact on the system
production rate, but may negatively affects the system yield. For
example, in robotic assembly the quality of the production process
is related to the robot repeatability and the output rate is related to
the robot speed. Robot repeatability deteriorates with the robot
speed [129]. This behaviour has been investigated in [186]. The
authors modelled multi-stage systems with quality-quantity
coupling machines. In these machines, the correlation between
efciency and yield is made explicit through an analytic relation.
The method supports the design of the optimal processing speed of
the machines in the system and has been applied to an automotive
case study [12].
The link between mix exibility and quality in exible
machining systems has also received attention in the literature.
A taxonomy for exible manufacturing systems is proposed in
[278]. A exibility evaluation toolbox in modern manufacturing
systems is addressed in [91]. Moreover, a method for assessing the
exibility of a manufacturing system, in an uncertain market
environment, under lifecycle considerations is developed in [7].
Part mix exibility provides to a system the ability of processing
different part types with relatively limited set-up times and
changeover costs. The level of exibility of a system affects the
product quality [161]. There are few examples showing that
system exibility is positively correlated with product quality
[232]. In [303] the author argues that exible modular assembly
systems support the achievement of higher product quality.
However, increased exibility can also deteriorate quality. For
example, consider a exible automotive paint shop [298]. When
shifting between different part batches characterized by different
colours a certain amount of defective parts that need to be
reworked is produced in the transient period since the colour is
contaminated by the one used for the previous batch. This
phenomenon is clearly strongly affected by decisions concerning
the set-up times and the job sequencing (see Section 3.2.5).
The effect of the design of recongurable systems on the
production logistics and quality performance has also received
attention in the past. [23,101] presented an approach for designing
system reconguration options according to a multi-criteria
decision making framework (Fig. 9). Starting from the analysis
of the product feature and demand requirements, and from a
database of available equipment modules a system-level tool
generates different potential reconguration alternatives. Their
KPIs are evaluated within a simulation environment and dominant
solutions are selected. ElMaraghy and Meselhy [80] proposed a

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

Fig. 9. Integrated approach for quality oriented design of RMSs [101].

framework to study the relation between maintainability and

quality in changeable manufacturing systems.
The impact of complexity in manufacturing and assembly
systems on several performance measures including quality and
production logistics metrics has been revised in [81,144].
Moreover, in [103] the impact of different plant complexity
sources on product quality was investigated based on the analysis
of real data from an automotive company. The results prove that
there exists a negative correlation between the number of chassis
produced in the plant and quality. Although system complexity has
many dimensions, product variety seems to be the most important
factor affecting production quality performance. In fact, the number
of product variants decreases the ability of learning from repetitive
operations and increases the probability of human errors. The link
between assembly system design for product variety and
performance was explored in [102]. Many papers address the
issue of quality and human induced errors in mixed-model
assembly systems [325]. Mixed-model assembly systems were
recognized as enablers for mass customization manufacturing.
However, highly proactive and knowledgeable workforce is
needed to effectively implement mixed-model systems in industry. In [2] the quality and productivity performance of mixedmodel assembly systems under human errors was evaluated. In
[271] it was reported that about 20% of the defects in the Fuji Xerox
China photocopy machine assembly systems was connected to
operators errors. It was the second cause for defects in the analysed
plant. Product and process related complexity metrics were
proposed to tackle this problem. The link between complexity
and performance measures in mixed-model assembly was also
algebraically analysed in [1].
3.1.2. Impact of process planning on quality
Manufacturing process planning is among the most knowledge
intensive decision-making activities undertaken in factories. In
this activity the product information is mapped on to the available
information for the various existing manufacturing resources to
determine a plan of action to convert the raw material into the nal
product. Process planning is normally carried out by a specic
human resource and depends on individual experience. Methodologies supporting process planners have been deeply analysed
since the 80s [42] and, over the last 10 years, a signicant number
of software tools focusing on Computer-Aided Process Planning
(CAPP) approaches have been developed [293]. CAPP systems use,
among others, articial intelligence methods to enable human
operators to select the most appropriate operations for
manufacturing. Currently the knowledge used within this activity
has been based on nominal models of manufacturing resources
[40,313]. While the nominal information pertaining to
manufacturing resource is static and does not change over time,
the capabilities of physical resources do, due to wearing of
mechanical components and tools. Capability proling [203] is a
method for recording these changes in the various capabilities of
manufacturing resources. With capability proling techniques, it is

781

possible to optimize the generation of the process plan to develop

solutions that are appropriate for the actual available hardware
and software rather than the nominal values. Capability proles
are generated by combining the nominal resource models with
actual values obtained from sensors on the shop oor and
predictive models.
In a production environment of resources with mixed capability
proles and varying reliability, both process planners and
schedulers tend to give priority to machines with more advanced
and unfailing services. All in all, this results in an uneven, distorted
load of these resources: while they are busy all the time, others are
idling. The throughput of such so-called exible job shops can,
however, substantially be improved if products are manufactured
via alternative routings. Nonaka et al. [208] presents a CAPP
method that, departing from the geometric product model and the
description of machining resources, generates a portfolio of
process plans with the objective to maximize the throughput.
The model is open to include quality related constraints, too. Next,
efcient load balancing and operation sequencing methods are
applied to schedule exible job shops by using the alternative
routings (i.e. process plans) for producing the same product. The
method that maximized a workshops throughput proved to be
robust and applicable even in large-scale industrial scenarios
[209]. The generation of alternative process plans is also the main
objective of the Network Part Program (NPP) approach [87] and of
non-linear process planning, in general. Non-linear process
planning goes beyond the static and strictly sequential nature of
traditional process plans that are often carried out without
considering the manufacturing system information [139]. The
idea of network part program is to delete non-technological
constraints from among the operations, transforming the sequential part program into a network of operations, each one
characterized by a set of G-M instructions. For instance, the part
program for the machining of a pallet can be easily built and rebuilt according to the workpieces that are really mounted on the
xture as a consequence of changes in demand mix and quantity
[218]. In case of unavailable resources part programs can be easily
adjusted and eventually split on different machine tools.
A rst attempt to develop NPP on industrial scale was led in the
Italian national project NetPP [21] where the approach was limited
to the production of pallets mounting one single part type on one
work area. Non-linear process planning able to support the
conguration of multi-xtures (pallet) with different parts has
been later developed [217] for managing small batches and a high
number of product variants. Currently, 12 installations of the NPP
are available in Europe. A process planning approach based on
network part program has been developed in the DEMAT EU
project [64] for a manufacturing system composed of ultra-light,
eco-compatible and energy efcient machine tools. Other European projects, such as ENEPLAN [83], analyse and propose non-linear
process planning techniques for hybrid processes, such as milling,
turning and laser cutting. One of the key challenges while applying
the NPP in industry is the need for specic procedures to provide a
quality certication of the entire Network Part Program, considering all the process path alternatives, instead of only certifying one
specic part program, as typically done with the traditional G-code
part programs.
3.1.3. Inspection planning in multi-stage systems
Inspection planning deals with the denition of the part quality
inspections in the production system and with the denition of the
multi-sensor system for process monitoring. Both technologies
serve as data gathering systems to feed SQC, Statistical Process
Control (SPC) and Condition Based Maintenance (CBM) procedures
with useful information to perform a machine and process state
diagnosis and the implementation of corrective or preventive
actions to restore in-control manufacturing system behaviour. This
diagnosis-oriented strategy focuses on the near-zero level of defect
generation. Here, the part quality inspection plans feed product
quality assurance and the consequent activation of defect

782

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

management strategies, including scrap, rework and repair. These

strategies allow smoothing the defect propagation throughout
process stages and to the nal customer. A review of the most
advanced automatic inspection and process data gathering
technologies is provided in Section 4. The use of these technologies
for complex product validation is revised in [180]. While product
inspection allocation techniques have been revised since 1980
[225], less attention has been given to process sensor distribution
strategies. Concerning part inspection planning, two major tasks
have to be solved:
 Inspection characteristics identication and analysis. The necessary
inspection characteristics have to be identied and analysed at
each process stage.
 Inspection process conception and allocation. According to the
identied and analysed inspection characteristics, inspection
strategies have to be developed, which dene the test
procedures, cases and resources and align the inspection steps
to the test sequence in multi-stage production systems.
The rst task is essential for the overall success of inspection
planning, since all characteristics which are neglected might cause
fatal damages to tools, personnel, products or customers and, on
the contrary, unreasonable inspections cause inefcient test steps
and increased process complexity. A common consequence of
wrong characteristics identication in the planning phase is the
occurrence of No-Fault-Found failures made visible during the use
phase of the product [179,220,221]. These are in-tolerance failures
due to unexplored interactions during the process/inspection
planning phases, performed without taking into consideration
process capability proles. Hence, [242] introduced the concept of
perceived quality, which provides methodologies to identify and
measure customer demands, and add the requirements from
different product stakeholders in order to develop a holistic
product and system specication [11,243]. The risk assessed
specications are the input for the second major task of inspection
planning, i.e. the inspection strategy planning and execution. This
phase entails:







Determination of the point in time of inspection (when?).

Determination of the proper technologies for inspection (how?).
Determination of the inspection extend (how much?).
Determination of the inspection location (where?).
Determination of the inspection personnel (who?).
Selection of the inspection equipment (whereby?).

Although heavy interdependencies between the inspection

planning steps do exist [246], scientic approaches mainly focus
on the optimization of single inspection planning tasks. These
works mainly address the inspection extend problem, against
economical KPIs using statistical methods [82,125].Van Volsem et
al. [287] derives an algorithm for the cost-optimal inspection
extend, place, type and amount of inspection stations. The lack of a
holistic consideration of all inspection planning tasks was
addressed by Schmitt et al. [250] where the model was extended
to calculate business cases based on the risk attitude of the
inspection management. With this idea, the optimal solution is the
one that maximizes the decision makers value of benet.
Concerning the area of sensors allocation for process monitoring only few recent contributions are available. In a manufacturing
process, sensor distribution involves the determination of: (i) the
workstations at which to place the sensing devices; (ii) the number
of sensors required at individual stations; (iii) the location of
sensors within individual stations. Three major types of problems
have been considered in the literature [178]: (i) for a given
number of sensors, nd the optimal sensor locations; (ii) nd
the minimal number of sensors as well as the corresponding
locations; and (iii) given the distribution of q sensors, where to
distribute additional s sensors. These formulations lead to a
constrained non-linear optimization problem [71,108]. Wang and

Nagarkar [302] proposed a two-level hierarchical approach to

solve problems (i) and (ii) simultaneously.
Part quality inspection and process sensor planning have a
strong impact on the production logistics performance of the
manufacturing system. It has been shown by [238] that, for a
production line with 15 machines, the effective production rate of
the system if inspection stations are poorly allocated can be 15%
lower than the one corresponding to a good allocation of the same
number of inspection stations. As investigated in [55] three
fundamental phenomena determine this effect. Firstly, if a critical
product feature is remotely monitored, a quality information
feedback delay is generated. If dedicated inspection stages are
designed, i.e. each critical product quality characteristic is
measured by a dedicated inspection device, local monitoring
should be adopted. However, in order to save equipment costs and
to increase the inspection system life-cycle, recongurable and
exible inspection technologies have been recently proposed
[15,137] which are able to adapt to and to measure a set of product
features. In this case, remote monitoring is inevitable.
Secondly, the part inspection interferes with the cycle time of
the system, while process monitoring activities typically do not.
Therefore, a more extensive product inspection provides more
accurate information about the product quality but decreases the
total production rate of the system. Thirdly, as it will be discussed
in Section 3.2.3, the implementation of defect management
strategies affects the system dynamics and its performance. An
algorithm to allocate inspection stations in order to maximize the
throughput of conforming parts, considering the effect of these
three phenomena under predetermined inspection technologies
and tasks, has been proposed in [188]. Moreover the concept of
quality bottleneck in a system in addition to the traditionally
investigated productivity bottleneck concept has been formulated
in [296]. A quality bottleneck is a stage in a multi-stage system that
more severely affects the system yield [185]. Identifying quality
and productivity bottlenecks is an important activity for prioritizing sensor and part quality inspection distribution. However, more
extensive research on inspection and sensor planning for production quality targets should be developed, jointly taking into account
all the aforementioned aspects.
3.1.4. Quality control planning in multi-stage systems
In multi-stage systems the design of an effective and costefcient quality control strategy is of critical importance. For
recent reviews of quality control planning methods in multi-stage
systems see [262,283]. The major challenges that have been
tackled by researchers in this area include multi-stage variational
propagation modelling for quality control, process monitoring, and
root cause identication for multi-stage systems. The rst area will
be revised in Section 3.2.1. Concerning process monitoring for
multi-stage systems, SPC is the main technique used in practice for
quality and process monitoring. Control charts are the most
commonly adopted tools. However, most conventional SPC
techniques treat the multi-stage system as a whole and lack the
capability to discriminate among changes at different stages [192].
To overcome this problem, multivariate control charts based on
principal components and partial least squares analyses seem
attractive for multi-stage systems. More recently, some specic
SPC techniques have been developed to exploit the detailed
structure of multi-stage systems to achieve high detection power
and diagnostic capability. For example, an exponential weighted
moving average scheme has been proposed as a monitoring
method for multi-stage systems [312,326]. In the SPC area, after a
process change is detected, the diagnosis of root causes is left to
human operators. Signicant progress has been made towards
intelligent root cause diagnostics. These methodologies can be
roughly classied as (i) statistical-estimation-based methods [72]
and (ii) pattern-matching-based methods [175]. Both methods are
based on mathematical models that link the system error and the
system quality measurements. As a matter of fact, the majority of
available SPC approaches tackle the quality control planning

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

problem by selecting the optimal control chart parameters

(sampling frequency, sample size and control limits) with respect
to an economic objective function. Multiple criteria including
production logistics and maintenance performance are usually
neglected. According to [262] the complexity of multi-stage
systems requires a holistic system-level approach for effective
quality control. By intermeshing and linking closed-loop quality
control systems at various levels of the company unambiguous
rules for decisions at engineering and organizational levels emerge
(Fig. 10).

Fig. 10. Cascaded closed-loop quality control systems [249].

The continuous alignment between actual and target state

enables continuous improvement to be institutionalized in the
company. In this direction Wiendahl introduced the characteristic
curves in order to describe production ows based on the
bottleneck theory and theory of constraints (TOC) [307]. The
underlying idea is to jointly perform quality control planning at the
facility and process control levels. In order to improve the
transparency of quality control planning in multi-stage systems,
the method Quality Value Stream Mapping [96], can be used to
develop an optimal conguration of quality control along the
process chain [148]. By means of Quality Value Stream Mapping
the occurrence of defects, the effective integration of inspection
stations as well as the design of quality control loops can be
systematically visualized, analysed and improved.
3.1.5. Personnel allocation in multi-stage systems
The human factor has a fundamental role in achieving the
required production quality performance of a manufacturing
system [308]. The human element is considered as a key factor
in all the discussed company functions, i.e. production, quality and
maintenance. Root cause analysis and nal product verication
still mainly ground on humanly driven operations, also in highly
automated contexts such as the automotive industry. As a matter
of fact, all traditional quality improvement programs, such as the
World Class Manufacturing, ground on the attitude of workers
towards problem solving and waste elimination. Moreover,
corrective and preventive maintenance procedures require highly
skilled personnel to be performed in compliancy with the target
times and cost requirements. Furthermore, complex manufacturing and assembly tasks still entail manual operations in almost all
industrial sectors. Even the human-robot interaction paradigm,
that is currently under investigation and testing [138], stresses the
importance of the role of humans in advanced manufacturing
systems for performing non-repetitive assembly tasks. Due to
these implications between the workforce organization and the
operational performance of a plant, the allocation and management of personnel in manufacturing systems have motivated
signicant amount of work in the past. Sterman [270] showed that
the production quality strategy xed by the company strategic goals
may activate virtuous (work wiser) or vicious (work harder) loops
in the behaviour of the workforce towards these targets, depending
on the vision imposed by the management. From an operational
point of view, the problem of allocating maintenance personnel in
complex multi-stage systems has been investigated. Automated
ow lines where human operators are allocated to cope with

783

machine breakdowns and other tasks such as inspection, support

and control have been considered. In these systems, machine
failures and consequent repair actions play a dominant effect on
the performance. To cope with machine failures, a repair crew is
usually dedicated to the line. However, in order to save operating
costs, the repair capacity is generally limited and the repair crew
availability can be a performance bottleneck for the whole line. In
other words, when a failure occurs and all operators are busy in
repairing other stations, the machine is forced to wait before a
repair intervention is started, queued with other contingent
maintenance requests. This kind of machine idleness is known in
literature as interference [300] and the related problem is known as
the Machine Interference Problem (MIP) [269]. A literature survey
on methods to solve MIPs can be found in [98]. An advanced
approach to iteratively solve this problem has been proposed in
[140]. The author modelled the original system as two interacting
systems the rst being the automated ow line and the second
being the repair crew system. This approach has been applied to a
real engine block production line at Scania in [45], showing great
operational benets for the plant obtained by optimally allocating
repairmen to stations. Moreover, the effect of mixed workforce
skill levels has been analysed in [48]. In [75] the problem of
distributing an available repair effort in the system, considering
the impact on the system dynamics have been solved by an
analytical approach. Currently, approaches that include quality
considerations in these frameworks are not available. From a
quality point of view, an approach to optimally design SPC control
chart parameters, also considering the limited manpower, is
proposed in [311]. Another area that has received attention in the
last years is the analysis of the impact of the workforce behaviour
on the operational performance of the system. In [106] the impact
of workers absenteeism on the performance of assembly lines is
investigated. The authors found that specic cross-training
strategies may reduce the loss of performance due to absenteeism.
However, cross training may reduce quality [108,182]. Furthermore, the impact of workforce learning on the system performance
improvement during the ramp-up phase has been analysed in [67].
Although preliminary approaches exist, the analysis of the impact
of workforce on the production quality performance of a
manufacturing system is a relatively new research area where
the main challenges is the difculty in developing reliable models
of the human behaviour.
3.1.6. Advanced integrated business models for production quality
Business model innovation is a relatively new concept in the
manufacturing industry. Traditionally, innovation in this sector
was primarily based on technology innovation. In the last decade,
industry competitiveness has been stained by the increasing
turbulence of the market. To face this situation, companies were
motivated to innovate their business models towards the
establishment of long-term relationships with their customers,
and the provision of value-added services beyond the technical
ones [306]. The topic of proposing new business models for
machine tool builders has attracted the attention of researchers
and industrialist only in recent years. The concept of Recongurable Manufacturing Systems [18] paved the way to the idea of
establishing a collaborative relation between producer and
customers to design and manage the system over its life-cycle.
Later in [258] the idea of delivering services for the system
adaptation and enable module re-use for different customers was
proposed. A strategic approach for developing such services has
been suggested [256]. Moreover, a CIRP Collaborative Working
Group Industrial Product Service Systems (IPS2) [187] has been
launched, with the objective of investigating benets and
operating modes for implementing the product-service idea in
B2B relations. The implementation of such concepts is the core idea
of the EU funded project RobustPlaNet [229]. In this project,
guidelines to select the best business model and collaboration
mechanism depending on the stakeholders situation are proposed.
In recent years full-service contracts and reliability warranties

784

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

have widely spread in almost all kinds of business, starting with

the aerospace and defence industry. Today automotive OEMs for
instance make contracts with their equipment suppliers for a
period of up to 10 years xing maintenance costs and performance
gures.
Within the new product-service oriented business models the
suppliers benet by gathering detailed data on how their machines
perform in real-case application in the global eld. Moreover, if
additional warranties on the quality of parts produced by the
equipment are integrated, detailed statistics on defects and root
causes are made available [148]. This is made possible by remote
monitoring systems implemented on the machines [193]. For
example, such a service is provided by Mori-Seiki when they
remotely monitor, and if possible, maintain, their CNC machine
tools via mobile communication networks worldwide (in 2011,
almost 6000 machines). As declared by the company, this service
affects not only the availability of resources, but also product
quality and resource efciency. In this context, a company
delivering the product-service will be able to increase the
protability of the business only by considering the quality,
productivity and maintenance aspects under an integrated view.
Indeed, any inefciency in one of these aspects may result in a
penalty and a value loss in the service provision. Therefore, the
contracts regulating the implementation of these new business
models and the production quality targets should be designed in a
coherent and non-conicting way.
3.1.7. Supply-chain design for production quality targets
To design a supply chain from a production quality point of view,
a detailed understanding of the failure propagation, the behaviour
of individual nodes in the chain and the overall tolerance
management are required. Within the German AiF Project iQ.net
a multi-agent simulation approach was developed to nd best
congurations of networks towards an integrated quality target
systems [19,146]. A similar problem was addressed in [196] where
the performance and viability of centralized and decentralized
production networks, under heavy product customization, were
investigated.
One of the major challenges in managing production quality in
globalized and highly distributed supply chains is the distribution
of production quality targets to a multitude of suppliers distributed
worldwide, each one having specic process capabilities and
production management strategies and goals [266]. A signicant
example is provided by Wiendahl, who studied the production
network of a German company producing weighing systems [309].
The competitiveness of the company was based on a solid
standardization of the modules composing the product and a
rationalized design of the variant differentiation. The production
quality standards were achieved by manufacturing high addedvalue components in production sites characterized by highly
skilled personnel and highly capable systems and to dedicate to the
less capable sites the production of low technological content
parts. This leaning affected the way the production system was
designed in the different sites, requiring dedicated solutions in the
early stages of production and exible solutions in the product
customization stages.
The problem is even more complex if differentiated products
dedicated to markets with specic, location-dependent production
quality requirements are considered. This is a growing trend in
globalized production networks, due to the rapid demand growth
in BRICS countries. To cope with this challenge the idea of frugal
products has been developed [183]. Frugal product innovation is
the process of removing nonessential features from a durable good,
such as a car or phone, in order to sell it in developing countries.
According to [3] in the automotive industry about 90% of
components are globally standardized while only 10% are adapted
to specic market requirements. On the contrary, in consumer
goods, these gures are almost reversed. Therefore, specic
advantages of the location are to be incorporated into the product
and production design processes, such as site-specic conditions

regarding production technologies and capabilities. From a

production quality point of view this translates into additional
burden on the design process. Indeed, not only the local production
process capabilities have to be considered but also the locationdependent product quality specications have to be met. Although
several works have addressed and formalized this challenge [195]
approaches to support decision making in this context have not
been developed.
Another relevant phenomenon, which reduces the ability of
meeting production quality targets, is the generation of obsolete
components caused by poor information exchange between
stakeholders in presence of highly customized products and
unexpected change of demand. These obsolete components are
excess inventory for the suppliers, which ultimately result in parts
to be scrapped and recycled. This problem is in the core of many
manufacturer-supplier relations where the parties have asymmetric information about the demand and cost items and should share
not only the benets but also the risks of operating the channel
[290]. Advanced technologies and cooperation-oriented contracts
can help reducing the impact of this phenomenon. For example, in
[197] a method of dynamically querying supply chain partners to
provide real time or near real time information regarding the
availability of parts required for the production of highly
customizable products is developed. This method utilizes Internet-based communication and real time information from RFID
sensors. The feasibility of this approach is demonstrated with its
implementation in a typical automotive case. The implementation
of advanced information technologies in supply chain for
improving product quality is also stressed in [290,313]. Egri and
Vancza [77] surveyed short-term two-echelon supply channel
coordination methods and presented a decentralized version of the
newsvendor model where the parties have the right incentive for
sharing their private information. The decision on production
quantities is in the hand of the supplier whose rational decision
concurs with the overall optimum. Hence, local decisions based on
asymmetric information coordinate the channel. Further work
resulted in a methodology for designing decentralized coordination protocols for supply chains where autonomous partners
operate in an asymmetric information setting, under the burden of
some uncertainty [78,288,289].
3.2. Operational, control and management phase
3.2.1. Multi-stage quality correlation analysis
Modern multi-stage manufacturing processes typically involve
processing and assembly stages whose output quality is signicantly affected by the output quality of preceding stages in the
system [171]. In multi-stage manufacturing processes, understanding how a defect generated in a specic process stage
propagates to the next process stages and what effect this
propagation has on the nal product quality is a complex task.
Two major types of correlations can be found in multi-stage
manufacturing systems:
 Quality correlation: the quality of the product processed at a given
stage is highly dependent on the quality of the output at specic
upstream processes.
 Failure correlation: the propagation of a defect generated in
upstream processes causes machine or tool integrity problems in
downstream processes, such as increased degradation and tool
wear, or sudden tool breakage, or process instability.
The most diffused market available Six-Sigma and SPC software
tools do not analyse stage correlations in multi-stage systems, but
concentrate on single process monitoring, control and optimization. Engineering methods and advanced Multivariate Statistical
Process Control (MSPC) methods have been proposed to model and
monitor correlations in multi-stage processes.
Of the engineering methods, SOVA (Stream of Variations
Analysis) has been proposed for assembly systems and machining

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

785

process-chains [261]. This approach is based on a state-space

representation of the correlation between the product deviations
at consecutive process stages whose structure is driven by
engineering knowledge about the processes and whose coefcients are tuned by KPC (Key Product Characteristics) measurements at the different stages [70] (Fig. 11). A detailed description
and review of SOVA model with applications to quality control
for multi-stage manufacturing processes is presented in [34].
Applications of the SOVA model to predict the propagation of
variation in the assembly are found in [29,28]. The SOVA model has
also been used to determine adjustments in multi-station
assembly processes [73].

Fig. 12. Representation of a manufacturing/assembly system with SPC and

corrective maintenance [279].

Fig. 11. Main characteristics of the Stream of Variation Analysis.

In contrast, advanced MSPC methods are based on elaborating

KPCs measurements to nd the parameters of simplied multiple
process statistical models [162]. In other words, they do not a priori
assume any specic structure of this correlation. Both approaches
support only quality correlation modelling, but do not offer the
possibility of eliminating defects at downstream correlated stages,
as a mean to achieve zero defect propagation.
These methods overlook the logistics analysis of the product
ow throughout the process stages. A recent contribution [210]
goes in the direction of overcoming these limitations. The authors
empirically studied the correlation between the occurrence of
machine failures and the quality problems in the produced parts.
The authors concluded that this correlation is the main cause for
quality problems in the analysed semiconductor manufacturing
fab (tsunami effect) and used this result for bottleneck identication. This paper gives a clear idea of the potential industrial
benets of methodologies addressing this problem.
Variation modelling in correlated serial-parallel multi-stage
systems has also been studied. The main problem in this context is
the presence of multiple variation propagation modes in the
production run when process routes vary from part to part. For a
practical example, refer to the automotive and semiconductor
cases reported in Section 1.2. If the different multiple process
routes share at least one workstation, the SOVA approach proposed
in [104] is applicable to model variation propagation. If there are
multiple variation propagation modes, only MSPC approaches can
be applied for modelling and monitoring [118].
3.2.2. Integrated methods for production quality prediction
Grounded on manufacturing system engineering background,
integrated models and analysis tools for predicting quality and
productivity performance measures in manufacturing systems
have been proposed. These models integrate product features and
specications, process out-of-controls, typical logistics parameters, such as machine failures and limited capacity buffers, and
corrective maintenance into a unique framework. The representation of a system including these aspects [279] is presented in
Fig. 12. Most of the methodological contributions in this area are
focused on serial production lines. The existing approaches
mainly differ in the type of operational and quality failures and
quality control mechanisms they model. In [185] the performance
of systems where quality failures exhibit Bernoulli-type behaviour and no correlation exists between consecutive parts are
analysed. However, in production systems Markovian-type outof-controls, where the quality of parts is dependent on the

machine state, are commonly observed. In [131], an analytical

method to estimate the yield and the total and effective
production rates of asynchronous lines having machines subject
to Markovian quality failures is proposed. When in normal
operating conditions the machine does not produce any defective
item; after transition to the quality failure state occurs, the
machine produces only defective products. The quality control
action is modelled through a transition that forces the machine to
shift to an un-operational state, for the repairing process.
This transition is considered to be xed and is taken as input,
therefore no link to specic quality control practices is assumed.
The authors later extended the approach to longer production lines
in [132] by proposing a decomposition approach. In [52] and [54],
approaches to evaluate the performance of synchronous
manufacturing lines, considering quality control actions triggered
by statistical control charts are proposed. The previous binary
quality failure model was replaced by considering multiple
operational states, each one characterized by its specic fraction
of defective parts produced. As in the previous contribution, the
control mechanism is modelled through a transition probability
that takes the machine to an un-operational state. However, this
probability is analytically derived by combining the control chart
and the system parameters. In [41] a queuing model was proposed
to investigate in detail the link between SPC delay and ow time in
a system with inspection stations. The applications of these
methods have led to the identication of interesting phenomena
due to the trade-off between quality and production logistics
performance and to the derivation of design and operational
decisions to exploit this trade-off. As emerges from this analysis,
currently the research integrating production logistics and quality
aspects has only concentrated on rst order performance measures
of the system, while due-date performance measures are
neglected. Although manufacturing system oriented methods for
estimating the variance of the cumulated output have been
proposed [49], they do not consider the implication with the
quality performance. A recent review of these approaches can be
found in [275]. The only study on quality robustness in
manufacturing system is proposed in [157] where Bernoulli
production lines are investigated.
3.2.3. System level defect management policies
Defect management policies in manufacturing systems can be
activated after the defect has been detected by part inspection
technologies or visual inspection performed by human operators. A
framework for characterizing defect management policies in
manufacturing systems has been proposed in [44]. These policies
include scrap, in-line rework, off-line rework, repair, and
downgrade. The implementation of these policies affects the
overall system performance in terms of quality, production
logistics and maintenance KPIs. Therefore, the selection of the
most proper defect management policy for each specic defect

786

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

type should be taken under an integrated production quality view. A

method to quantitatively predict the production quality implication
of these policies is shown in [44], where the approach is applied to
the electric drive assembly system described in Section 1.2. The
impact of scrap policies on the manufacturing system performance
has been widely studied in the literature. Helber [99] and Han and
Park [97] proposed decomposition methods for the performance
evaluation of multi-stage production lines in which the fraction of
non-conforming parts is scrapped by the system. In these
contributions, the production of non-conformities is a random
event and it is not related to the machine state. Colledani and Tolio
[53] analysed the performance of a manufacturing system with offline inspection stations and scrapping of defective items. In this
work, the fraction of defective parts at the scrapping stage is a
function of the rate of occurrence of a process out-of-control, of the
system dynamics and of the sampling frequency at the inspection
stations. In general, scrapping positively affects the system yield at
the cost of reducing the total throughput of a system and of loosing
the workpiece value accumulated until the stage where scrapping
takes place. Scrapping causes waste. Therefore, suitable waste
management policies have to be dened in view of the zero-waste
manufacturing paradigm. Recently, an attractive option for
industrial waste management is the principle of industrial
symbiosis. According to this principle, the waste of a company is
transformed into raw materials for another company. The EU
funded project ZERO-WIN [323] addresses this challenge and
proposes several demonstrators in different sectors including
electronic products, buildings, automotive and consumer goods.
Rework policies have also been extensively investigated. The rst
works in this area only considered the implications of rework with
the total production rate of the system, neglecting the impact on
the product quality. In [155] the performance of the systems with
rework are analysed through an overlapping decomposition
approach. Defective parts are extracted from the main ow,
undergo preparation processes in off-line stages and then are reinserted upstream in the line for reprocessing. This method is
applied to a real paint shop in [156]. In [31] a model of a
manufacturing system with multiple rework loops is developed.
More recently, the impact of rework on the quality and
productivity performance has been jointly analysed. In [158] a
method to evaluate rst time quality and quality buy rate in a
rework/repair paint shop in the automotive industry is proposed.
This work has been extended to include a model of the part
variations along the process stages [120]. In [20] the authors
analysed the performance of production systems monitored by online SPC and rework of defective products. They considered for the
rst time the dependency of the rework probability on the quality
control system parameters and on the machine deterioration
dynamics. The application to a real system in the white good
industry is reported. Product repair has attracted less attention in
the literature. Repair differs from rework since re-processing is not
applied. Repair can be performed manually or automatically in the
same stage where the defect is generated or downstream at
correlated process stages. For example, in [43] repair strategies for
restoring the quality of electric drives are discussed from a
technical feasibility point of view. Product repair strategies have
also been applied in manufacturing systems producing batteries
for electric cars [121].
3.2.4. Integrated production control and quality control strategies
Pull-type production control policies have been developed with
the objective of coordinating different stages in manufacturing
systems to react to actual occurrences of demand rather than
future demand forecasts like in push systems, according to the JIT
principles. These policies include kanban, basestock, CONWIP
policies. For a recent review and a comparison of these policies see
[167]. These mechanisms control part releases at the stages in
order to resolve the trade-off between unsatised demand and
holding costs. A pull mechanism explicitly controls the WIP in the
system. Other dynamic production control policies, based on

indirect WIP control activated by workstation processing rate

adjustments, are proposed in [14,176,281,130]. These policies have
been traditionally compared in terms of production logistics
performance and due-date performance. However, since the
inventory has a relevant impact on quality, as commented in
Section 3.1.1, production control policies affect production quality
performance. A comparison between quality and productivity
performance in pull and push systems is discussed in [165]. In [57]
the effective production rate of a CONWIP controlled production
system where machines are monitored by SPC is analysed. A
CONWIP population level that maximizes the effective throughput
was found for any closed-loop system conguration, including
those providing a atness region in the total production rate curve.
In [55] an optimization method for jointly setting kanban card
levels and the parameters of statistical control charts has been
proposed. The objective is the maximization of an income function,
obtained by combining both production and quality related costs
and the constraint include the satisfaction of a target production
rate of the system. The authors showed that the optimal solution
obtained by the joint approach has an income that can be 40%
higher than the impact obtained by isolated approaches. Other
approaches have tackled the problem from a quality-oriented
perspective. Del Castillo [63] investigated how the parameters of a
statistical control chart affect the service level of a small
production-inventory system with stochastic demand by proposing a semi-Markov model of the system. In [259] a productioninventory-inspection system is analysed modelling a specic
portion of a front-end semiconductor facility comprising an
etching tool. The product is defective if the dust level caused by
the etching tool increases a xed value. An optimal part release
policy from the production station into the buffer is developed to
reduce the number of inspected items under a given defect risk. In
[301] the problem of batch release control in the semiconductor
industry is taken into account. In detail, incoming parts are
selected in order to minimize the within batch variability. Then, for
each specic batch, the production parameters at the downstream
stages are adjusted in order to reduce the between-batch
variability observed at the end of the line. Selective assembly
can be also seen as an integration of quality and production
control. In the literature, quality-oriented approaches have been
developed that study the effect of the component sorting policy on
the assembled product quality [184]. Other studies investigate the
effects of process adaptation on the performance of selective
assembly systems [126]. Process adaptation means that the
nominal value of the key quality characteristics of the component
produced with the more capable production process can be
adjusted according to pre-dened states of the system, in order to
increase the component matching. Recently, simulation and
analytical methods have been used for predicting the impact of
specic adaptation policies on the production quality performance
[174]. An attempt to apply selective assembly to address the part
gap control problem in automotive remote laser welding applications is done in the EU project RLW Navigator [228]. Selfoptimizing, advanced cognitive methods have been proposed for
tolerance matching problems. A cognitive architecture for
advanced production and quality control systems was set up by
Schmitt et al. [248,251,252] based on the generic human cognitive
process model. An essential requirement upon cognitive architecture is to build a model of the production processes to be optimized
and to be equipped with learning capability.
3.2.5. Set-up and batching for quality and productivity
In a exible, multi-product manufacturing system, set-ups are
detrimental for the system production rate as they increase the
system unproductive time. Substantial research has been devoted
to product sequencing considering set-ups [22]. Mathematical
programming [222] and soft computing techniques [90] are
usually used to nd the optimal sequence in multi-product
contexts. Recently, a real time control strategy has been proposed
for production control in exible manufacturing with set-ups

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

based on an extension of the hedging point policy, called hedging

zone policy [284]. All these works neglect the implications with
product quality. Set-ups are usually considered as detrimental to
product quality, especially in those contexts where a ramp-up
operation is required after the set-up has been performed. This
phenomenon has been studied in [298] where the authors
developed a Markovian approach to predict the fraction of
defective parts produced in a exible system producing in batch
with set-ups. This work is extended to product sequencing
problems with quality considerations [297]. Methods to identify
the quality bottleneck sequence in exible manufacturing systems
are developed in [296] and [299] based on data collected from the
shop oor. In [172] an approach to set-up planning for ensuring the
achievement of desired quality specications is proposed. A part
variation model is developed and dynamic programming is used
for the optimization. The authors showed that currently adopted
experience-based approaches tend to be conservative and allocate
work to the most capable processes, since a model of the
variational patterns is not available. Frequent machine set-ups
among different part types also generate small production runs
(batches) and this may be disruptive for product quality since
process learning through data analysis is prevented [116]. The area
of quality control and process improvement in small-batch
production has been recently investigated. Small batches can
benet very little from statistical inference built exclusively on
work piece-related dimensional data. Therefore, correlation and
behavioural patterns that link machine-process related status
parameters with more general dimensional and shape-related
metrological parameters are needed [58]. To this purpose, multisensor data fusion has been applied (see Section 4.3). Another
promising technology to address this problem is prole monitoring
where functional features instead of dimensional features are
monitored [194]. All these methods have been traditionally
evaluated in terms of quality responsiveness performance. The
implications with production logistics and maintenance are
typically neglected. Another non-negligible implication of setups needs to be taken into account. A higher number of set-ups and
lower batch size decrease the WIP. Therefore, by the decreasing the
WIP they positively impact on product quality, for two reasons.
With small batches the quality information feedback can be
propagated with short delays, thus enabling a more reactive
control of the system. Moreover, if goods are perishable, small lots
are benecial. [30] formulates a mixed-integer non-linear programming model for optimal lot splitting to account for possible
effects of production run length on product quality in cellular
manufacturing systems. The main idea is that when a production
lot is split into several alternative routes, the production run in
each route will be shortened and this may result in better product
quality. Production planning and lot sizing with variable and
uncertain yield has also been addressed. Yield decrease causes the
need for larger lots to be able to deliver to the customer the
required quantities of conforming items, given that a certain
fraction will be defective [285]. For a review of rst works in this
area see [318]. In [36] a lot sizing model with quality and
maintenance considerations is proposed. Sarker et al. [236]
developed models for an optimal batch quantity for a manufacturing system with rework of defects. A closed-form formula for the
system performance is developed and non-linear optimization
techniques are used for deriving the optimal lot size.
3.2.6. Maintenance in multi-stage systems
A multi-stage system is a multi-components system built on
different interactions between components by considering that the
states of components inuence the states of others. From a
dependability point of view, inuences are materialized by the
principle of failure interaction or stochastic dependency [205].
Traditionally, the performance of manufacturing systems under
corrective maintenance operations has been estimated. Corrective
maintenance means that the maintenance activity is only activated
after a failure has been realized. For a review of methods for

787

throughput estimation in unreliable manufacturing systems under

corrective maintenance see [159]. For a review of methods
focusing on assembly systems see [100]. More recently, the impact
of preventive maintenance policies, including both Age Based (ABM)
and Condition Based (CBM), on the system productivity performance has been considered [115]. A key issue in analysing the
impact of preventive maintenance is the modelling of degradation
processes. For instance, degradation can be due to the wear of
tools, xtures, or machine components. Degradation is a progressive process that increases the probability of breakdown over time.
A degrading process has been usually modelled as a homogenous
or inhomogeneous Markov chain with increasing failure rate [320].
In practice, in order to select the number of operational states and
the elements of the transition rate matrix that better approximate
a real degradation process two approaches can be coupled. Firstly,
a transition from one operational state to another may correspond
to a specic physical event and the states to a specic machine,
component, or tool condition. For example, in [264] a multi-state
degradation model for a friction drilling process subject to tool
wear was developed. Secondly, the states can be thought of in an
abstract way as representing a discretization of the deterioration
process. In this way, generally distributed degradation processes
can be modelled by using acyclic continuous Phase Type (PH)
distributions [213]. Such process and equipment degradation
models have been incorporated in manufacturing lines [47,316]
and assembly systems [234] in order to approximate the
productivity performance of the entire line. These models have
shown that, in multistage systems, optimal single stage, isolated
maintenance policies can be weak if applied to multi-stage
systems due to the locations of the bottlenecks in the system [10].
In multi-stage systems, when a component is failing, one or
several components can be impacted and require to be maintained.
Indeed inspection or replacement action on one component can
initiate, at the same time, another action on a dependent
component, as advocated by Block Replacement Policy (BRP). In this
sense, the downtime to repair a component is an opportunity to
maintain other components in the system. This opportunity is
offering an additional solution space to the conventional planning
and scheduling of maintenance activities for anticipating failures,
reducing system unavailability and maintenance costs (both direct
and indirect). Such opportunity can be detected by prognostics
tools, which allow assessing the degradation state and the
performance level of each component, also taking into account
the dependencies and the dynamics of the whole system [204].
Opportunity is the central concept of Opportunistic Maintenance, initially dened in [181] as Opportunistic Replacement
and Inspection Policy. For a recent review see [24]. Opportunistic
Maintenance may entail (i) grouping several maintenance actions
together [69,100], (ii) associating a preventive maintenance task to
a corrective maintenance action [16,237], or (iii) performing a
maintenance action during an opportunity [6,114,133,154].
Opportunistic Maintenance can be implemented only when
technical and economic conditions are satised in a way to
achieve optimal maintenance in terms of a balance between
maintenance cost and component/system reliability.
3.2.7. Joint maintenance and production scheduling
The problem of planning maintenance and production operations in manufacturing systems has been widely addressed by
researchers, mainly in isolation. The survey by Wang discusses the
major contributions of the maintenance policies of deteriorating
systems [294]. These contributions and methods do not take into
account any other system information, such as inventory levels or
the states of the other resources in the system. Therefore
opportunistic maintenance policies cannot be considered. In
the production planning area, machine failures are usually
considered, but without having any control over the failure
occurrence. Indeed, optimal production policies often model only
two-state Markovian failure mechanisms, which means that
component lifetime is exponentially distributed, thus precluding

788

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

preventive maintenance with increasing failure rate from being

modelled [319]. Recently, the idea of jointly planning production
and maintenance has received attention in the literature
[39,267,268]. The common features of these studies are that they
examine the joint optimal production and maintenance policy for
a machine and an inventory that decouples the facility with the
stochastic market demand. In [33] the value of the integration of
production and maintenance in planning and scheduling is a 30%
reduction in mean job tardiness. Jin et al. [119] proposed an
option-based model for a joint production and maintenance
planning system to avoid backlog in case of non-deterministic
demand. Kaihara et al. [124] proposed a model for the optimal
maintenance and production planning in re-entrant lines.
Moreover, [317] analysed the benets of adjusting the throughput
of degrading machines on the maintenance scheduling efciency.
The idea is that the throughput of a resource in highly degraded
states can be adjusted in order to control the remaining life before
the next maintenance operation. The authors investigated the joint
control of machine reconguration and maintenance in a parallelserial manufacturing system by simulation [324]. Other works
addressing simultaneous planning of production and maintenance
are [5,35,92,211]. This integration is studied also in [4,62,127] with
considerations related to deterioration processes with increasing
failure rate. In particular, in [68] a buffered two stage system where
the rst stage goes through degradation was considered. A Markov
decision model was proposed to optimally select the critical state of
the rst stage that activates preventive maintenance, for each buffer
level. The rationale is that when the buffer is close to empty,
maintenance should be activated only in very critical degradation
states [145,286].
3.2.8. Joint maintenance and quality control strategies
In all the previously revised works, degradation only entails and
increased chance of a failure to happen. However, degradation of a
component/system is one of the major factors that cause defective
product output [142]. In that way, one conventional solution to
reduce the number of defective units, is to conduct preventive
maintenance strategy on the component/system to keep it in
acceptable conditions, in phase with requirements expected on the
product quality. Another solution is to sample the output to screen
out the defective units. An innovative way is to combine these two
approaches in order to integrate these two management practices
(e.g. Preventive Maintenance (PM) and SQC) for nding the optimal
policy minimizing the total expected cost. The combined application of SQC techniques and PM methods for achieving higher
product quality and more effective use of resources has been
investigated at single machine level [32,321]. Later [170,265]
combined the two approaches, at system level. In [142] an optimal
adaptive control policy for machine maintenance and product
quality control is derived. Moreover, [215] developed an optimal
process control and maintenance procedure under general
deterioration patterns, and [37] minimized the cost of an
integrated systemic approach to process control and maintenance
based on EWMA control charts by using genetic algorithm. A
performance measurement system for integrated SPC and CBM
procedures is proposed in [122]. These works show that quality
control based on product measurements can be useful for
enhancing improved maintenance procedures.
3.2.9. Joint maintenance, production logistics and quality control
The interactions between quality, production logistics and
maintenance have been mainly analysed at managerial, tactical
and strategic levels [147]. Only few recent works have proposed
quantitative tools and methods to address operational issues with
integrated approaches. In [9] a conceptual model for planning
maintenance operations in view of the overall plant effectiveness,
including quality and productivity considerations, and protability
was proposed. Its application to a Swedish paper-mill plant
showed that an extra prot of about 8 million SEK per year could be
achieved. In [128] a regression model linking quality and

maintenance hours to productivity performance is proposed for

the food industry. The authors show that signicant correlations
exist between these aspects. Pandey et al. [216] proposed an
integrated planning model to improve the performance of a single
stage system. Firstly, the selection of the optimal maintenance
interval and process quality control parameters is jointly
performed. Then the optimal preventive maintenance interval is
integrated within the production planning problem in order to
determine the optimal batch sequence that will minimize penaltycost incurred due to schedule delay. The results show that cost
savings up to 80% can be achieved. Radhoui et al. [224]
investigated the full integration of these areas under a different
perspective in a single stage system. They proposed to use the
fraction of defective parts delivered by the system as the
monitored variable that can activate maintenance activities if a
xed threshold is exceed. Then, a buffer of nite size is located. The
parameters of this control system are jointly optimized. Dhouib et
al. [66] proposed an hedging point policy approach to control a
degrading quality machine, where a security stock of nished
products is maintained in order to respond to demand and to avoid
shortages during maintenance actions. These works extend the
approaches revised in Section 3.2.8 to the case of imperfect quality.
The only work integrating all aspects in a multi-stage model has
been proposed by Colledani and Tolio [56]. The authors developed
a model of a multi-stage asynchronous serial line where machines
are subject to deterioration. While going through deteriorated
states, increasing failure rate and decreasing yield are observed.
For this machine, a maintenance action based on SPC or PM or both
can be carried out. The authors showed that in multi-stage
systems, while selecting the optimal maintenance thresholds, the
solutions obtained by neglecting quality deterioration and the
system dynamics are always sub performing in terms of effective
production rate and always overestimate the length of maintenance cycles. Moreover, the optimal maintenance policy at one
machine is highly affected by the parameters of the other machines
in the system. The benets in the effective production rate can be
as high as 30%.
4. Enabling technologies
The improvement of production quality targets in industrial
processes requires the development of new and emerging
technologies. This section revises advanced technology and ICT
solutions supporting the production quality target.
4.1. Product inspection technologies
The implementation of the production quality paradigm requires
advanced technologies for on-line data gathering, incorporating:
 3D exible part verication through integration of multi-sensor,
multi-resolution systems.
 ICT architectures to support in-line inspection and data sharing
at system level.
Manufacturing of complex 3D parts for highly customized small
batch production is creating a strong demand of advanced
inspection systems to characterize the physical aspects of the
produced parts. During the last years, efforts was made to increase
the availability of affordable point-based measurement systems,
capable of acquiring large amounts of data as point coordinates,
with high accuracy. Although very accurate in measurements,
these technologies are very difcult to be implemented on-line due
to the invasive interference with the process cycle time. To
overcome these limitations, 3D inspection analysis has progressed
signicantly, where advanced sensors have evolved from single
sensors into multi-sensors [17,88].
Multi-sensors have several advantages: (i) different inspection
technologies can be used; (ii) the number and type of sensors are
not limited; (iii) diverse data can be added adaptively; and (iv)

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

multi-scale data can be merged into a single multi-scale model.

Today, sensors are classied as contact and non-contact types
based on the interaction with the inspected part. A typical multisensor conguration includes both contact sensors (e.g. touch
probes) and non-contact sensors (e.g. video cameras, laser
scanners and micro-probes) mounted on the same machine. The
multi-sensor head of the Nikon scanning system [206] is depicted
in Fig. 13. Contact and non-contact sensors each have their own
working principles and properties, simultaneously providing
diverse and complementary data that can considerably improve
inspection. Customers continue to demand smaller, faster, cheaper,
easier and more precise metric solutions. Multi-sensors can meet
these demands more effectively than can multiple single-sensors
[173]. Contact sensors usually provide sparse and very accurate
high-resolution (HR) data with long inspection time, while noncontact sensors [226] provide dense and less accurate lowresolution (LR) data but can measure thousands of points per
second. Due to their differing accuracies, these two data sets can be
regarded as multi-resolution data. However, the majority of multisensor data is often lost and unutilized. Therefore, the main
challenge lies in how to utilize the LR data despite its lower
accuracy.

Fig. 13. Multi-sensor head: laser scanner (left) and touch probe (right).

789

work piece during machining [305]. When this intimate level of

integration between machining and checking is attainable, it is
usually called in process control, to differentiate it as a special
case of in-line control. For a variety of reasons, machining such as
turning or milling do not yet easily afford in-process control,
though they certainly give room to off-process, in-line measurement. However, still there is chance to monitor machine critical
parameters (e.g. spindle stray vibrations, spindle torque and axial
force, lubricant pressure, tool integrity and in-y kinematics,
instantaneous power proles, etc.) as the machining occurs. A
review on techniques for sensor monitoring of machining
operations is presented in [277]. The scheme of a sensorized
turning machine provided by Artis GmbH, a Company of Marposs
Group, is represented in Fig. 14. However, correlating process
signals and product geometry metrological data is still a challenge
in manufacturing operations.

Fig. 14. Multi-sensor system for process monitoring.

The automation of such inspection technologies inline poses

interesting challenges to data acquisition. Large information
volume acquired through multiple advanced sensors will have
to be processed, fused, organized and stored in the data repository
so that it can be used both on-line and off-line, at any stage of the
product life-cycle where it is requested. The use of MTConnect, the
open software standard for data exchange and communication
between manufacturing equipment recently launched by the
Association for Manufacturing Technology [292] as well as other
communication standards [207] can potentially support this task.
Such information-sharing platform would enable interoperability
among (i) the different heterogeneous sensors distributed in the
process chain and among (ii) the data gathering system and the
data processing models and methods that have been developed for
the production quality paradigm.
Moreover, the availability of on-line data gathering technologies will support the achievement of dynamic inspection planning
decisions, according to the real state of the processes and
resources. One of the rst attempts in this direction was done by
[227] with the Productivity+ tool. This tool implements process
probing checks and dynamically modies the inspection process
plan. This solution can support maintenance decision making at
shop oor level. This aspect is a fundamental step towards the
implementation of integrated quality, maintenance and production logistics tools.

4.3. Multi-sensor data fusion technologies

Multi-sensor data fusion technology is usually applied to
metrology data [304] and signal data [84]. With reference to
metrology data, more frequently companies face the problem of
inspecting Geometric Product Specications (GPS) of complex and
freeform surfaces [110], rather than simple dimensions. Currently,
a trade-off between resolution and inspection time is dened per
single-sensor processes. Most of the approaches for data and
signals fusion from multi-sensors consider AI-based tools (e.g.
neural networks) that usually need long training times [166].
Current data fusion approaches lack in the ability of coping
simultaneously with metrology data and signal data, and do not
refer to the correlation between them [59].
Multi-resolution modelling has been explored by a variety of
algorithms over the years [190,276]. The underlying idea behind all
these algorithms is to adapt the resolution to the features level-ofdetail. Multi-resolution methods can be applied to a variety of
engineering problems such as reconstruction of 3D models from
cloud of points [13,263], reconstruction from 2D images acquired
by Scanning Electron Microscope (SEM) [235], and verication of
3D freeform parts with noise. After the 3D model is reconstructed,
it can be analysed utilizing multi-scale nite element methods
[219] or optimized applying domain decomposition re-meshing
techniques [322].

4.2. Process monitoring technologies

4.4. Learning technologies and cognitive computing methods.

The production quality paradigm needs technologies to provide

higher degree of machine-condition awareness and advanced
diagnostics and maintenance ability with lower interference with
the system production rate.
There are specic machining processes, e.g. precision grinding,
centerless grinding, that may offer superior accuracy and stability
because of their own inherent properties combined with the
possibility to check the machine parameters and to measure the

Since most of the methods revised in Section 4 rely on resource

reliability and degradation models obtained from eld data,
learning technologies and cognitive computing methods are
relevant for the production quality scope. Intelligent data analysis
and classication methodologies have been proposed in the last
years in order to predict behaviours of machines and processes and
to provide fault diagnosis based on predictor variables. A
comprehensive review of these approaches can be found in

790

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

[277]. Thus, taking into account the results and the conclusions
achieved from such methods, knowledge extraction and decision
making support tasks can be accomplished with the aim of
reinforcing holistic quality system and suggest actions to be
performed in order to maintain the resources in the system. There
exist several recent techniques to deal with this issue. The most
important include Decision and Regression Trees, Classication
Rules, Fuzzy Models, Genetic Algorithms, Bayesian Networks,
Articial Neural Networks. Failure detection and classication are,
in general, well established and accomplished nowadays [199];
however, prognosis procedures based on multiple conditions are
not yet well dened [85].

maintenance procedures. In [273] the most promising production

monitoring tools are revised. In particular, the andon system has
been recognized as an important enabler for quality and
maintenance operations. The andon system is one of the elements
that make up the principle of jidoka. It consists in an intuitive
stage visualization system with a suitable human machine
interface (HMI). Recent works have quantitatively investigated
the impact of andon on production, quality and maintenance
performance [160], showing benets at system level. Production
monitoring systems can be also support system performance
monitoring [223] and bottleneck identication [163], which are
fundamental tools for production improvement.

4.5. E-maintenance technologies

4.8. ICT and digital manufacturing technologies

Internet and tether-free technologies contributed to a transition

from traditional fail and x maintenance practices to a predict
and prevent e-maintenance approach [151,198]. Through
e-Maintenance relevant data, information, knowledge, and intelligence is made available and usable at the right place, at the right
time for anticipated maintenance decisions. E-Maintenance is a
holistic enterprise process which integrates the principles already
implemented by tele-maintenance [150] into web-services and
collaboration platforms, thus encompassing traditional synchronization principles [113]. These systems are supported by technologies such as wireless communications, mobile components (e.g.
Personal Digital Assistant, SmartPhones, Graphic tablets, harden
laptops), smart sensors, MEMS, Global Positioning System (GPS), and
Web CMMS. E-maintenance is offering services/processes and tools
to monitor the asset degradation, to diagnose its root cause and to
prognosticate its remaining useful life in order to optimize the asset
utilization in the facility. The performance assessment and
prediction tools can also be used with links to Product LifeCycle
Management (PLM) applications [95,291]. For example, [74]
proposes the concept of Watchdog Agent as an infrotronics-based
prognostics approach for product performance degradation assessment and prediction. Yang and Lee [315] develops a Bayesian Belief
Network to investigate the causal relationship among process
variables on the tool and evaluate their inuence on wafer quality in
semiconductor manufacturing.

ICT solutions can support the production quality principle by

vertically transferring process, quality control and diagnostics data
at decision making level and vice versa, by achieving interoperability and integration of multi-scale and heterogeneous shop oor
data and by avoiding defect generation in manual operations
through virtual and augmented reality tools.
As concern the rst issue, an ICT architecture to support
visualizing and managing interactions among different quality and
process control loops at shop oor level has been developed [247].
The concept of the software combines the assessment of a quality
control loop and its step-by-step improvement. The tool can
additionally be used to efciently guide the user through all the
steps that are required for dening a new quality control loop to be
implemented. A Man, Machine, Maintenance and Economy
(MMME) software tool has been developed and applied to FIAT
in [8]. It enables to vertically connect maintenance data to
production and business data for improved decision making.
Concerning the second requirement, recently knowledge-based
virtual platforms have been developed for enabling data exchange
and interoperability among heterogeneous ICT tools for factory
planning, reconguration and management. For example, in the
Virtual Factory Framework (VFF) EU funded project the application of a semantic data model for virtual factories to support the
design of manufacturing systems is proposed [123]. An ontologybased framework can be used to share consistent design and shopoor data between different heterogeneous software tools
including, 3D virtual environments, discrete event simulation
models and analytical models, at both process and system levels.
Similarly, Industrie 4.0 is an initiative focused on Cyber-Physical
Systems (CPS) with approaches to opportunistic maintenance,
self-sensing and self-conguring components and plug-andproduce manufacturing.
Concerning the support of complex manual operations,
augmented reality solutions have been proposed for improved
assembly [177] and maintenance tasks [202,212]. These technologies proved to be effective in different business cases in reducing
the defect generation during assembly tasks, in limiting the time,
and increasing the capabilities in complex repair and maintenance
operations. Moreover, they have been suitably applied for operator
training programs in manufacturing [153].

4.6. Product traceability technology

In complex manufacturing system layouts featuring parallel
machines, non-linear material ows and split-merge stages
tracking the product throughout the process-chain stages and
correlating its features to the specic processes that manufactured/assembled the product is a priority for improved maintenance, quality and logistics control. The introduction of product
data into the conventional control system ensures the arrival of the
correct items for manufacturing and to trace the product (and its
subcomponents) through the different stages of system. Product
traceability through RFID technologies has been proposed as a
solution to this problem [253]. With this technology traceability
and quality error management can be performed. Direct tracking of
the items moving through the system enables accurate status of
each item to be maintained in a suitable data store. This provides
correlation information that can be used with any error that is
detected, enabling simpler root cause analysis and fault diagnostics [256]. Moreover, defect management policies implemented
at shop oor can benet from this technology, as specic
knowledge of the process sequence can improve product rework
and repair operations.
4.7. Production monitoring technologies
Production monitoring technologies at shop oor level can
support the production quality paradigm by providing the required
set of data about the timed sequences of the states of the resources
in the system to enable the joint production, quality and

5. Future research priorities

The classication of methodologies and the results illustrated in
Sections 3 and 4 have been used to identify regions of the
interaction model where methodologies are still missing and
attention by the research community is required.
Proactive on-line defect repair policies. The traditional belief that
stage correlation raises a problem for control in multi-stage
systems should be drastically changed into an opportunity for
improvement. Indeed, if the result of a downstream process stage
is affected by the incoming product quality, then the downstream
process stage can have an impact on the incoming product quality,
and, if properly controlled, can possibly correct a defect generated
in the upstream stages.

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

Joint analysis of quality correlations and system dynamics. Quality

variation propagation in correlated multi-stage systems and
production logistics performance of manufacturing systems have
been typically analysed as two independent areas. However, in
correlated multi-stage manufacturing systems, there are chances
of integrating these models to provide a comprehensive and
integrated model of the quality and equipment failure propagation
dynamics at system level.
Preventive maintenance to improve quality robustness. The
quality robustness of a system with machines having out-ofcontrols have only been tackled under simple machine reliability
models (single state model). Preventive maintenance not only
enables high service levels but it also affects the quality of the parts
produced. If properly controlled, preventive maintenance operations can thus reduce the variance of the output, thus increasing
the service level. Therefore models are required to jointly consider
preventive maintenance and quality robustness.
Production quality in complex system architectures. The production quality problem has been investigated only in relatively simple
manufacturing system architectures. However, modern systems
frequently feature resource by-pass strategies, re-entrant ows,
and process multiple part types in the same system. At present,
only the production ow performance of these systems has been
investigated. Re-entrant lines have been studied in [253,257,314].
Multiple part types systems have also received attention [46], as
well as systems with split and merge operations [50]. However,
there is a lack of comprehensive models that can consider
production quality issues in complex system architectures combining many of the features described above.
Production quality in advanced material ow control policies.
Similar to the previous case, there is a need to analyse production
quality in systems with non-FIFO dispatching and sequencing
rules. The advent of agent-based systems [191], intelligent product
principles and autonomous control methods have proposed realtime decisions for dispatching parts to the available resources, to
increase the system resilience to disturbances [26,254,255].
However, the implications of this complex management rules
on the product quality and on the degradation of resources is at
present almost unexplored.
Multi-level, multi-scale modelling, simulation and analytics for
production quality. Capturing the interactions between the
manufacturing and assembly process layer, where the defects
are generated, and the system layer, where the defects are
propagated, seems to be a promising potential solution to move
towards a balanced design and management of manufacturing
systems for production quality. The coupling of these layers into a
comprehensive modelling platform could support the selection of
process parameters in view of the overall production quality system
performance as well a the selection of the optimal system
architecture and part routing considering different process
capabilities. However, this would require the proper integration
of multi-physics, multi-scale models of manufacturing processes
and systems that are currently designed as isolated tools.
Formalized data structures and interaction mechanisms among
maintenance, quality and production departments. One of the major
challenges to be solved to achieve high production quality is the lack
of formalized data structures integrating product quality, resource
maintenance and productivity areas. These data are commonly
collected in separated databases by departments that rarely share
these data. Moreover, these departments do not share company
control objectives. This situation tends to generate conicts among
the competing production quality elements instead of privileging
the search for a negotiated, overall balanced solution. Therefore,
the company structure, the management control systems, and the
ICT infrastructure should be re-designed and aligned, possibly with
the help of ontologies, to reach production quality goals.
Dynamic control of production quality in the system life-cycle. To
achieve a proper co-evolution level between product, process and
system life-cycles, statistics on production quality targets should
be more extensively collected. Moreover, transitions among

791

different production, quality control strategies should be planned

with an integrated view of the problem. In particular, although
preliminary research has been carried out in this direction, the
transient analysis of production quality performance in
manufacturing systems ramp-up needs to be further investigated
to develop suitable technologies and methodologies to reduce the
costs and times of system ramp-ups.
New strategies and business models for production quality. It is well
known in the manufacturing strategy literature that the companys
strategy, the business model, and the operational performance need
to be perfectly aligned in order to gain competitiveness in the
market. The new production quality paradigm needs to be supported
by a specic manufacturing strategy and business model. The
alignment between maintenance and manufacturing strategies has
been recently recognized as a key enabler for competitiveness in
modern manufacturing industries [231]. Traditional quality management models tend to incorporate the zero-defect vision by
maximizing the overlap of customer demands and delivered product
features, while costs have to be minimized. The potentials of
production quality is in this way heavily reduced via two implied
restrictions: (i) market-sided assumption, i.e. companies cannot
decide about their target production quality level; (ii) organizationalsided assumption, i.e. the companies processes contain all the skills
to operate exactly according to the desired product features. A new
model has to allow companies to identify strategic targets and
balance them towards their desired equilibrium.
An entrepreneurially inuenced understanding of quality
management should be dened as the immediate and waste-free
fullment of customer demands under consideration of the
strategic objectives, the conditions and the actual companys
resources/skills. In this line, the Aachen Quality Model [241,245],
Fig. 15, takes strategic objectives, the entrepreneurial conditions
and the companys capabilities into account.

Fig. 15. Aachen quality model for entrepreneurial quality.

6. Concluding remarks
This paper has formalized the basic hypothesis of a new
production quality paradigm for modern, zero-defect oriented,
manufacturing industries. This new paradigm relies on a strong
interaction among production logistics, quality and maintenance
functions. The major interactions among production logistics,
quality and maintenance variables have been formalized and
mapped into a model that can represent a practical tool to support
companies to characterize signicant trade-offs in their plants. The
most advanced methodologies and enabling technologies facilitating the implementation of this new paradigm in industry have
been revised and directions for future research have been
provided. The production quality paradigm represents a valuable
opportunity for modern manufacturing organizations and, at the
same time, a challenge calling for the development of new
advanced knowledge-based manufacturing models and tools.
Acknowledgements
The authors would like to thank Dr. Steven Peters from KITwbk, Dr. Andrea Ratti, Dr. Anteneh Teferi and Eng. Sara Costa from

792

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

Politecnico di Milano for the support in the preparation of the

paper, and Prof. Olav Egeland from NTNU, Prof. Darek Ceglarek
from University of Warwick, Prof. Peter Wiendahl, Prof. Hoda
ElMaraghy, Prof. Waghui ElMaraghy from University of Windsor,
Prof. Jack Hu and Prof. Yoram Koren from University of Michigan,
Mr. Jan Aichele from Robert Bosch GmbH, Mr. Kepa Aiesta and
Mr. Xabier Bernarte from Gamesa, Mr. Ziprani from Marposs
Group, and Mr. di Cecio and Mr. Camanzi from ENKI Srl for the
useful suggestions and comments.

References
[1] Abad AG, Guo W, Jin J (2013) Algebraic Expression of System Congurations
and Performance Metrics for Mixed Model Assembly Systems. IIE Transactions
46(3):230248.
[2] Abad AG, Paynabar K, Jin JJ (2011) Modeling and Analysis of Operator Effects
on Process Quality and Throughput in Mixed Model Assembly Systems.
Journal of Manufacturing Science and Engineering 133(2):19.
[3] Abele E, Meyer T, Naher U (2007) Global Production: A Handbook for Strategy
and Implementation, Springer.
[4] Aghezzaf E, Najid NM (2008) Integrated Production Planning and Preventive
Maintenance in Deteriorating Production Systems. Information Sciences
178(17):33823392.
[5] Aghezzaf EH, Jamali MA, Ait-Kadi D (2007) An Integrated Production and
Preventive Maintenance Planning Model. Operational Research 181(2):679685.
[6] AlDurgam MM, Duffuaa SO (2013) Optimal Joint Maintenance and Operation
Policies to Maximize Overall Systems Effectiveness. International Journal of
Production Research 51(5):13191330.
[7] Alexopoulos K, Mourtzis D, Papakostas N, Chryssolouris G (2007) DESYMA:
Assessing Flexibility for the Lifecycle of Manufacturing Systems. International
Journal of Production Research 45(7):16831694.
[8] Al-Najjar B, Jacobsson M (2013) A Computerised Model to Enhance the CostEffectiveness of Production and Maintenance Dynamic Decisions: A Case
Study at Fiat. Journal of Quality in Maintenance Engineering 19(2):114127.
[9] Alsyouf I (2007) The Role of Maintenance in Improving Companies Productivity and Protability. International Journal of Production Economics
105(1):7078.
[10] Ambani S, Li L, Ni J (2009) Condition-Based Maintenance Decision-Making for
Multiple Machine Systems. Journal of Manufacturing Science and Engineering
131(3):31009.
[11] Amini P, Schmitt R (2013) Analyzing the Deviation of Product Value Judgment. Smart Product Engineering 23rd CIRP Design Conference, 765775.
[12] Arinez J, Biller S, Meerkov SM, Zhang L (2010) Quality/Quantity Improvement
in an Automotive Paint Shop: A Case Study. IEEE Transactions on Automation
Science and Engineering 7(4):755761.
[13] Azernikov S, Miropolsky A, Fischer A (2003) Surface Reconstruction of
Freeform Objects from 3D Camera Data Based on Multiresolution Volumetric
Method. ASME Transactions Journal of Computing and Information Science in
Engineering 3(4):334338.
[14] Bai SX, Gershwin SB (1995) Scheduling Manufacturing Systems with WorkIn-Process Inventory Control: Single-Part-Type Systems. IIE Transactions
27(5):618624.
[15] Barhak J, Djurdjanovic D, Spicer P, Katz R (2005) Integration of Recongurable
Inspection with Stream of Variations Methodology. International Journal of
Machine Tools and Manufacture 45(4):407419.
[16] Bedford T, Dewan I, Meilijson I, Zitrou A (2011) The Signal Model: A Model for
Competing Risks of Opportunistic Maintenance. European Journal of Operational Research 214(3):665673.
[17] Bernard A, Fischer A (2002) New Trends in Rapid Product Development. CIRP
Annals Manufacturing Technology 51(2):635652.
[18] Bi ZM, Lang SYT, Shen W, Wang L (2008) Recongurable Manufacturing
Systems: The State of the Art. International Journal of Production Research
46(4):967992.
[19] Book J, Kumar M, Lanza G, Srai J (2012) Quality Risk Management in Global
Supply Networks: An Agent-Based Approach. Proceedings of The Annual Cambridge International Manufacturing Symposium, 140.
[20] Borgh D, Colledani M, Simone F, Tolio T (2007) Integrated Analysis of
Production Logistics and Quality Performance in Transfer Lines with Rework.
Proceedings of the 6th International Conference on Analysis of Manufacturing
Systems, 1520.
[21] Borgia S, Pellegrinelli S, Petro` S, Tolio T (2014) Network Part Program
Approach Based on the Step-NC Data Structure for the Machining of Multiple
Fixture Pallets. International Journal of Computer Integrated Manufacturing
27(3):281300.
[22] Boysen N, Fliedner M, Scholl A (2009) Sequencing Mixed-Model Assembly
Lines: Survey, Classication and Model Critique. European Journal of Operational Research 192(2):349373.
[23] Bryan A, Ko J, Hu SJ, Koren Y (2007) Co-Evolution of Product Families and
Assembly Systems. CIRP Annals Manufacturing Technology 56(1):4144.
[24] Budai G, Dekker R, Nicolai RP (2006) A Review of Planning Models for Maintenance and Production, Economic Institute Report, 2006-44.
[25] Bulgak AA (1992) Impact of Quality Improvement on Optimal Buffer Design
and Productivity in Automatic Assembly Systems. Journal of Manufacturing
Systems 11(2):124136.
[26] Butala P, Sluga A (2006) Autonomous Work Systems in Manufacturing Networks. CIRP Annals Manufacturing Technology 55(1):521524.

[27] Callen JL, Fader C, Krinsky I (2000) Just-In-Time: A Cross-Sectional Plant

Analysis. International Journal of Production Economics 63(3):277301.
[28] Camelio JA, Hu SJ, Marin SP (2004) Compliant Assembly Variation Analysis
Using Component Geometric Covariance. Journal of Manufacturing Science and
Engineering 126(2):355360.
[29] Camelio J, Hu SJ, Ceglarek D (2003) Modeling Variation Propagation of MultiStation Assembly Systems with Compliant Parts. Journal of Mechanical Design
125(4):673681.
[30] Cao D, Defersha FM, Chen M (2009) Grouping Operations in Cellular
Manufacturing Considering Alternative Routings and the Impact of Run
Length on Product Quality. International Journal of Production Research
47(4):9891013.
[31] Cao Y, Subramaniam V, Chen R (2012) Performance Evaluation and Enhancement of Multistage Manufacturing Systems with Rework Loops. Computers &
Industrial Engineering 62(1):161176.
[32] Cassady CR, Bowden RO, Liew L, Pohl EA (2000) Combining Preventive
Maintenance and Statistical Process Control: A Preliminary Investigation.
IIE Transactions 32(6):471478.
[33] Cassady CR, Kutanoglu E (2003) Minimizing Job Tardiness Using Integrated
Preventive Maintenance Planning and Production Scheduling. IIE Transactions 35(6):503513.
[34] Ceglarek D, Huang W, Zhou S, Ding Y, Kumar R, Zhou Y (2004) Time-Based
Competition in Multistage Manufacturing: Stream-of-Variation Analysis
(Sova) Methodology Review. International Journal of Flexible Manufacturing
Systems 16(1):1144.
[35] Celen M, Djurdjanovic D (2012) Operation-Dependent Maintenance Scheduling in Flexible Manufacturing Systems. CIRP Journal of Manufacturing
Science and Technology 5(4):296308.
[36] Chakraborty T, Giri C, Chaudhuri S (2008) Production Lot Sizing with Process
Deterioration and Machine Breakdown. European Journal of Operational Research 185(2):606618.
[37] Charongrattanasakul P, Pongpullponsak A (2011) Minimizing the Cost of
Integrated Systems Approach to Process Control and Maintenance Model by
EWMA Control Chart Using Genetic Algorithm. Expert Systems with Applications 38(5):51785186.
[38] Cheng CH, Miltenburg J, Motwani J (2000) The Effect of Straight and U-Shaped
Lines on Quality. IEEE Transactions on Engineering Management 47(3):321334.
[39] Cheung KL, Hausman WH (1997) Joint Determination of Preventive Maintenance and Safety Stocks in an Unreliable Production Environment. Naval
Research Logistics 44(3):257272.
[40] Chin K-S, Zheng L-Y, Wei L (2003) A Hybrid Rough-Cut Process Planning for
Quality. International Journal of Advanced Manufacturing Technology 22(9):
733743.
[41] Chincholkar M, Herrmann W (2008) Estimating Manufacturing Cycle Time
and Throughput in Flow Shops with Process Drift and Inspection. International Journal of Production Research 46(24):70577072.
[42] Chryssolouris G, Chan S, Cobb W (1984) Decision Making on the Factory
Floor: An Integrated Approach to Process Planning and Scheduling. Robotics
and Computer-Integrated Manufacturing 1(3):315319.
[43] Colledani M, Coupek D, Aichele J (2013) Proactive Quality Control System for
Defect Reduction in the Production of Electric Drives. 3rd International Electric
Drives Production Conference and Exhibition, Nuremberg, Germany.
[44] Colledani M, Coupek D, Verl A, Aichele J, Yemane A (2014) Design and
Evaluation of In-Line Product Repair Strategies for Defect Reduction in the
Production of Electric Drives. Proceedings of the 24th CIRP Design Conference,
Milano, Italy.
[45] Colledani M, Ekvall M, Lundholm T, Moriggi P, Polato A, Tolio T (2010)
Analytical Methods to Support Continuous Improvements at Scania. International Journal of Production Research 48(7):19131945.
[46] Colledani M, Gandola F, Matta A, Tolio T (2008) Performance Evaluation of
Linear and Non Linear Multi-Product Multi-Stage Lines with Unreliable
Machines and Finite Homogeneous Buffers. IIE Transactions 40(6):612626.
[47] Colledani M, Gershwin SB (2013) A Decomposition Method for Approximate
Evaluation of Continuous Flow Multi-Stage Lines WITH General Markovian
Machines. Annals of Operations Research 209(1):540.
[48] Colledani M, Lundholm T, Ratti A, Tolio T (2012) Analytical Methods for
Performance Evaluation of Transfer Lines with Shared Repair Capacity at
Scania. VTI 2012 Vehicle Technology Integration Conference, 573578.
[49] Colledani M, Matta A, Tolio T (2010) Analysis of the Production Variability in
Multi-Stage Manufacturing Systems. CIRP Annals Manufacturing Technology
59(1):449452.
[50] Colledani M, Tolio T (2005) A Decomposition Method to Support the Conguration/Reconguration of Production Systems. CIRP Annals Manufacturing
Technology 54(1):441444.
[51] Colledani M, Tolio T (2006) An Analytical Method to Support the Design of
Production Systems Monitored by Statistical Process Control. CIRP Seminar on
Manufacturing Systems 423431.
[52] Colledani M, Tolio T (2006) Impact of Quality Control on Production System
Performance. CIRP Annals Manufacturing Technology 55(1):453456.
[53] Colledani M, Tolio T (2009) Performance Evaluation of Production Systems
Monitored by Statistical Process Control and Off-Line Inspections. International Journal of Production Economics 120(2):348367.
[54] Colledani M, Tolio T (2011) Joint Design of Quality and Production Control in
Manufacturing Systems. CIRP Journal of Manufacturing Science and Technology
4(3):281289.
[55] Colledani M, Tolio T (2011) Integrated Analysis of Quality and Production
Logistics Performance in Manufacturing Lines. International Journal of Production Research 49(2):485518.
[56] Colledani M, Tolio T (2012) Integrated Quality, Production Logistics and Maintenance Analysis of Multi-Stage Asynchronous Manufacturing Systems with
Degrading Machines. CIRP Annals Manufacturing Technology 61(1):455458.

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

[57] Colledani M, Yemane A (2010) Integrated Quality and Production Logistics
Analysis of Closed Loop Manufacturing Systems. IEEE Conference on Emerging
Technologies and Factory Automation, 18.
[58] Colosimo BM, Semeraro Q, Pacella M (2008) Statistical Process Control for
Geometric Specications: On the Monitoring of Roundness Proles. Journal of
Quality Technology 40(1):118.
[59] Colosimo BM, Senin N (2011) Geometric Tolerances: Impact on Product Design,
Quality Inspection and Statistical Process Monitoring, Springer.
[60] Daaboul J, Da Cunha C, Bernard A, Laroche F (2011) Design for Mass Customization: Product Variety vs. Process Variety. CIRP Annals Manufacturing
Technology 60(1):169174.
[61] De Jong A, Smit K (2012) Collaborative Contracts for Inter-Organisational
Quality Systems. Journal of Quality in Maintenance Engineering 18(2):171182.
[62] Dehayem Nodem FI, Kenne JP, Gharbi A (2011) Simultaneous Control of
Production, Repair/Replacement and Preventive Maintenance of Deteriorating Manufacturing Systems. International Journal of Production Economics
134(1):271282.
[63] Del Castillo E (1995) Relations Between Control Chart Design Variables and
Production Control. International Journal of Production Research 33(10):
27092721.
[64] DEMAT, De-materialized manufacturing systems, EU FP7. Grant agreement
246020.
[65] Demir L, Tunali S, Eliiyi DT (2012) The State of the Art on Buffer Allocation
Problem: A Comprehensive Survey. Journal of Intelligent Manufacturing
9(2):122.
[66] Dhouib K, Gharbi A, Ben Aziza MN (2012) Joint Optimal Production Control/
Preventive Maintenance Policy for Imperfect Process Manufacturing Cell.
International Journal of Production Economics 137(1):126136.
[67] Digiesi S, Kock AA, Mummolo G, Rooda JE (2009) The Effect of Dynamic
Worker Behavior on Flow Line Performance. International Journal of Production Economics 120(2):368377.
[68] Dimitrakos TD, Kyriakidis EG (2008) A Semi-Markov Decision Algorithm for
the Maintenance of a Production System with Buffer Capacity and Continuous Repair Times. International Journal of Production Economics 111:752762.
[69] Ding F, Tian Z (2012) Opportunistic Maintenance for Wind Farms Considering Multi-Level Imperfect Maintenance Thresholds. Renewable Energy
45:175182.
[70] Ding Y, Ceglarek D, Shi J (2002) Fault Diagnosis of Multistage Manufacturing
Processes by Using State Space Approach. Journal of Manufacturing Science and
Engineering 124(2):313322.
[71] Ding Y, Elsayed EA, Kumara S, Lu JC, Niu F, Shi J (2006) Distributed Sensing for
Quality and Productivity Improvements. IEEE Transactions on Automation
Science and Engineering 3(4):344359.
[72] Ding Y, Zhou S, Chen Y (2005) A Comparison of Process Variation Estimators
for In-Process Dimensional Measurements and Control. ASME Transactions
Journal of Dynamic Systems Measurement and Control 127(1):6979.
[73] Djurdjanovic D, Zhu J (2005) Stream of Variation Based Error Compensation
Strategy in Multi-Stage Manufacturing Processes. ASME Conference Proceedings 42231:12231230.
[74] Djurdjanovic D, Lee J, Ni J (2003) Watchdog Agent An Infotronics-Based
Prognostics Approach for Product Performance Degradation Assessment and
Prediction. Advanced Engineering Informatics 17(34):109125.
[75] Dogan-Sahiner E, Altiok T (1998) Planning Repair Effort in Transfer Lines. IIE
Transactions 30(10):867882.
[76] Duou JR, Sutherland JW, Dornfeld D, Herrmann C, Jeswiet J, Kara S, Hauschild
M, Kellens K (2012) Towards Energy and Resource Efcient Manufacturing: A
Processes and Systems Approach. CIRP Annals Manufacturing Technology
61(2):587609.
[77] Egri P, Vancza J (2012) Channel Coordination with the Newsvendor Model
Using Asymmetric Information. International Journal of Production Economics
135(1):491499.
[78] Egri P, Vancza J (2013) A Distributed Coordination Mechanism for Supply
Networks with Asymmetric Information. European Journal of Operational
Research 226(3):452460.
[79] ElMaraghy H, Schuh G, ElMaraghy W, Piller F, Schonsleben P, Tseng M,
Bernard A (2013) Product Variety Management. CIRP Annals Manufacturing
Technology 62(2):629652.
[80] ElMaraghy WH, Meselhy KT (2009) Quality and Maintainability Frameworks
for Changeable and Recongurable Manufacturing. Changeable and Recongurable Manufacturing Systems, 321336.
[81] ElMaraghy W, ElMaraghy H, Tomiyama T, Monostori L (2012) Complexity in
Engineering Design and Manufacturing. CIRP Annals Manufacturing Technology 61(2):793814.
[82] Emmons H, Rabinowitz G (2002) Inspection Allocation for Multistage Deteriorating Production Systems. IIE Transactions 34(12):10311041.
[83] ENEPLAN Energy efcient process planning system, www.eneplan.eu.
[84] Esteban J, Starr A, Willetts R, Hannah P, Bryanston-Cross P (2005) A Review of
Data Fusion Models and Architectures: Towards Engineering Guidelines.
Neural Computing and Applications 14(4):273281.
[85] Ferreiro S, Arana R, Aizpurua G, Aramendi G, Arnaiz A, Sierra B (2009) Data
Mining for Burr Detection (in the Drilling Process). Distributed Computing,
Articial Intelligence, Bioinformatics, Soft Computing, and Ambient Assisted
Living, Springer, Berlin, Heidelberg12641273.
[86] Fleischer J, Spath D, Lanza G (2003) Quality Simulation for Fast Ramp Up. 36th
CIRP International Seminar on Manufacturing Systems, 33/6.
[87] Fogliazza G (2004) How Information Technology Enables the Potential Flexibility and Re-Congurability of Highly Automated Manufacturing Plants.
IFAC International Control Problems in Manufacturing, 389404.
[88] Fraden J (2010) Handbook of Modern Sensors: Physics, Designs, and Applications,
4th ed. Springer, New York.
[89] From Recall to Prevention, SGS Consumer Information Bulletin, 11/15/2012.

793

[90] Gagne C, Gravel M, Price WL (2006) Solving Real Car Sequencing Problems
With Ant Colony Optimization. European Journal of Operational Research
174(3):14271448.
[91] Georgoulias K, Papakostas N, Makris S, Chryssolouris G (2007) A Toolbox
Approach for Flexibility Measurements in Diverse Environments. CIRP Annals
Manufacturing Technology 56(1):423426.
[92] Gharbi A, Kenne JP (2005) Maintenance Scheduling and Production Control of
Multiple-Machine Manufacturing Systems. Computers & Industrial Engineering 48:693707.
[93] Goch G, Knapp W, Hartig F (2012) Precision Engineering for Wind Energy
Systems. CIRP Annals Manufacturing Technology 61:611634.
[94] Guerriero F, Miltenburg J (2003) The Stochastic U-Line Balancing Problem.
Naval Research Logistics 50:3157.
[95] Gulledge T, Hiroshige S, Iyer R (2010) Condition-Based Maintenance and the
Product Improvement Process. Computers in Industry 61(9):813832.
[96] Haefner B, Kraemer A, Stauss T, Lanza G (2014) Quality Value Stream
Mapping. Procedia CIRP 17:254259.
[97] Han MS, Park DJ (2002) Optimal Buffer Allocation of Serial Production Lines with
Quality Inspection Machines. Computers & Industrial Engineering 42(1):7589.
[98] Haque L, Armstrong MJ (2007) A Survey of the Machine Interference Problem.
European Journal of Operational Research 179:469482.
[99] Helber S (1999) Performance Analysis of Flow Lines With Non-Linear Flow of
Material, Springer-Verlag.
[100] Hu J, Zhang L, Liang W (2012) Opportunistic Predictive Maintenance for
Complex Multi-Component Systems Based on DBN-HAZOP Model. Process
Safety and Environmental Protection 90(5):376388.
[101] Hu J (2006) Co-Evolution of Product Families and Manufacturing Systems.
Presentation at the Species WG meeting in Kobe.
[102] Hu SJ, Ko J, Weyand L, ElMaraghy HA, Lien TK, Koren Y, Shpitalni M (2011)
Assembly System Design and Operations for Product Variety. CIRP Annals
Manufacturing Technology 60(2):715733.
[103] Huang N, Inman R (2010) Product Quality and Plant Build Complexity.
International Journal of Production Research 48(11):31053128.
[104] Huang Q, Shi J (2004) Variation Transmission Analysis and Diagnosis of
Multi-Operational Machining Processes. IIE Transactions 36(9):807815.
[105] INCOSE, International Council on Systems Engineering (2010) System Engineering Handbook v.3.2, INCOSE, San Diego, USA.
[106] Inman RR, Blumenfeld DE (2010) Assembly Line Team Sizing with Absenteeism. International Journal of Production Research 48(22):65376558.
[107] Inman RR, Blumenfeld DE, Huang N, Li J, Li J (2013) Survey of Recent Advances
on the Interface Between Production System Design and Quality. IIE Transactions 45(6):557574.
[108] Inman RR, Jordan WC, Blumenfeld DE (2004) Chained Cross-Training of
Assembly Line Workers. International Journal of Production Research
42(10):18991910.
[109] Inman R, Blumenfeld E, Huang N, Li J (2003) Designing Production Systems
for Quality: Research Opportunities from an Automotive Industry Perspective. International Journal of Production Research 41(9):19531971.
[110] I SO TC 213 Dimensional and Geometrical Product Specications and
Verication.
[111] IEC/ISO62264-1 (2003) Enterprise-Control System Integration Part 1: Models
and Terminology, ISO Edition.
[112] Iung B, Levrat E, Crespo Marquez A, Erbe H (2009) Conceptual Framework for
e-Maintenance: Illustration by e-Maintenance Technologies and Platform.
Annual Review in Control 33(2):220229.
[113] Iung B (2003) From Remote Maintenance to MAS-Based e-Maintenance of an
Industrial Process. Journal of Intelligent Manufacturing 14(1):5982.
[114] Iung B, Levrat E, Thomas E (2007) Odds Algorithm-Based Opportunistic
Maintenance Task Execution for Preserving Product Conditions. Annals of the
CIRP Manufacturing Technology 56(1):1316.
[115] Iung B, Veron M, Suhner MC, Muller A (2005) Integration of Maintenance
Strategies into Prognosis Process to Decision-Making Aid on System Operation. CIRP Annals Manufacturing Technology 54(1):58.
[116] Jaber Y, Khan M (2010) Managing Yield by Lot Splitting in a Serial Production
Line with Learning, Rework and Scrap. International Journal of Production
Economics 124(1):3239.
[117] Jambekar AB (2000) A Systems Thinking Perspective of Maintenance, Operations, and Process Quality. Journal of Quality in Maintenance Engineering
6(2):123132.
[118] Jin R, Liu K (2013) Multimode Variation Modeling and Process Monitoring for
Serial-Parallel Multistage Manufacturing Processes. IIE Transactions
45(6):617629.
[119] Jin X, Li L, Ni J (2009) Option Model for Joint Production and Preventive Maintenance System. International Journal of Production Economics 119:347353.
[120] Ju F, Li J, Xiao G, Arinez J (2013) Quality Flow Model in Automotive Paint
Shops. International Journal of Production Research 51(21):64706483.
[121] Ju F, Li J, Xiao G, Huang N, Biller S (2014) A Quality Flow Model in Battery
Manufacturing Systems for Electric Vehicles. IEEE Transactions on Automation
Science and Engineering 11(1):230244.
[122] Juuso EK, Lahdelma S (2013) Intelligent Performance Measures for Condition-based Maintenance. Journal of Quality in Maintenance Engineering
19(3):278294.
[123] Kadar B, Terkaj W, Sacco M (2013) Semantic Virtual Factory Supporting
Interoperable Modelling and Evaluation of Production Systems. CIRP Annals
Manufacturing Technology 62(1):443447.
[124] Kaihara T, Fujii N, Tsujibe A, Nonaka Y (2010) Proactive Maintenance Scheduling in a Re-Entrant Flow Shop Using Lagrangian Decomposition Coordination Method. CIRP Annals Manufacturing Technology 59(1):453456.
[125] Kakade V, Valenzuela F, Smith S (2004) An Optimization Model for Selective
Inspection in Serial Manufacturing Systems. International Journal of Production Research 42(18):38913909.

794

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

[126] Kayasa MJ, Herrmann C (2012) A Simulation-Based Evaluation of Selective

and Adaptive Production Systems (SAPS) Supported by Quality Strategy in
Production. Procedia CIRP 3:1419.
[127] Kazaz B, Sloan TW (2012) The Impact of Process Deterioration on Production
and Maintenance Policies. European Journal of Operational Research
227(1):88100.
[128] Khan MR, Darrab IA (2010) Development of Analytical Relation Between
Maintenance, Quality and Productivity. Journal of Quality in Maintenance
Engineering 16(4):341353.
[129] Khouja M, Rabinowitz G, Mehrez A (1995) Optimal Robot Operation and
Selection Using Quality and Output Trade-Off. International Journal of Advanced Manufacturing Technology 10(5):342355.
[130] Kim JH, Dufe NA (2005) Design and Analysis of Closed-Loop Capacity
Control for a Multi-Workstation Production Systems. CIRP Annals
Manufacturing Technology 54(1):455458.
[131] Kim J, Gershwin SB (2005) Integrated Quality and Quantity Modeling of a
Production Line. OR Spectrum 27(2):287314.
[132] Kim J, Gershwin SB (2008) Analysis of Long Flow Lines with Quality and
Operational Failures. IIE Transactions 40(3):284296.
[133] Koochaki J, Bokhorst JA, Wortmann H, Klingenberg W (2012) Condition Based
Maintenance in the Context of Opportunistic Maintenance. International
Journal of Production Research 50(23):69186929.
[134] Koren Y, Heisel U, Jovane F, Moriwaki T, Pritschow G, Ulsoy G, Brussel HV
(1999) Recongurable Manufacturing Systems. CIRP Annals Manufacturing
Technology 48(2):527540.
[135] Koren Y, Hu SJ, Gu P, Shpitalni M (2013) Open-Architecture Products. CIRP
Annals Manufacturing Technology 62(2):719729.
[136] Koren Y, Hu SJ, Weber TW (1998) Impact of Manufacturing System Conguration on Performance. CIRP Annals Manufacturing Technology
47(1):369372.
[137] Koren, Y., Katz, R., 2003, Recongurable apparatus and method for inspection
during a manufacturing process, United States Patent 6567162.
[138] Kruger J, Lien TK, Verl A (2009) Cooperation of Human and Machines in
Assembly Lines. CIRP Annals Manufacturing Technology 58(2):628646.
[139] Kruth JP, Detand J (1992) A CAPP System for Nonlinear Process Plans. CIRP
Annals Manufacturing Technology 41(1):489492.
[140] Kuhn H (2003) Analysis of Automated Flow Line Systems with Repair Crew
Interference. Analysis and Modeling of Manufacturing Systems, Springer :
155179.
[141] Kumar U, Galar D, Parida A, Stenstrom C, Berges L (2013) Maintenance
Performance Metrics: A State-of-the-Art Review. Journal of Quality in Maintenance Engineering 19(3):3-3.
[142] Kuo Y (2006) Optimal Adaptive Control Policy for Joint Machine Maintenance
and Product Quality Control. European Journal of Operational Research
171(2):586597.
[143] Kurfer J, Westermeier M, Tammer C, Reinhart G (2012) Production of LargeArea Lithium-Ion Cells Preconditioning, Cell Stacking and Quality Assurance. CIRP Annals Manufacturing Technology 61(1):14.
[144] Kuzgunkaya O, ElMaraghy H (2006) Assessing the Structural Complexity of
Manufacturing Systems Congurations. International Journal of Flexible
Manufacturing Systems 18(2):145171.
[145] Kyriakidis EG, Dimitrakos TD (2006) Optimal Preventive Maintenance of a
Production System with an Intermediate Buffer. European Journal of Operational Research 168:8699.
[146] Lanza G, Book J, Kippenbrock K, Saxena A (2013) Innovative Quality Strategies
for Global Value-Added-Networks. Robust Manufacturing Control, 271286.
[147] Lanza G, Peters S (2012) Integrated Capacity Planning Over Highly Volatile
Horizons. CIRP Annals Manufacturing Technology 61(1):395398.
[148] Lanza G, Stricker N, Peters S (2013) Ad-Hoc Rescheduling and Innovative
Business Models for Shock-robust Production Systems. Procedia CIRP
7:121126.
[149] Lanza G, Sauer A, Kolmel A (2012) Conguration of a Multi-Use Battery
Production. CIRP 5th International Conference on Changeable, Agile, Recongurable and Virtual Production, 473478.
[150] Lee J (1998) Teleservice Engineering in Manufacturing: Challenges and
Opportunities. International Journal of Machine Tools & Manufacture
38:901910.
[151] Lee J, Ni J, Djurdjanovic D, Qiu H, Liao H (2006) Intelligent Prognostics Tools
and e-Maintenance. Computers in Industry Special Issue on e-Maintenance
57(6):476489.
[152] Lee WR, Beruvides MG, Chiu YD (2007) A Study on the QualityProductivity
Relationship and its Verication in Manufacturing Industries. The Engineering
Economist 52(2):117139.
[153] Leu MC, ElMaraghy HA, Nee AY, Ong SK, Lanzetta M, Putz M, Bernard A (2013)
CAD Model Based Virtual Assembly Simulation, Planning and Training. CIRP
Annals Manufacturing Technology 62(2):799822.
[154] Levrat E, Thomas E, Lung B (2008) Odds-Based Decision-Making Tool for
Opportunistic Production-Maintenance Synchronisation. International Journal of Production Research 46(19):52635287.
[155] Li J (2004) Performance Analysis of Production Systems with Rework Loops.
IIE Transactions 36(8):755765.
[156] Li J (2004) Throughput Analysis in Automotive Paint Shops: A Case Study.
IEEE Transactions on Automation Science and Engineering 1(1):9098.
[157] Li J, Blumenfeld DE, Marin SP (2008) Production System Design for Quality
Robustness. IIE Transactions 40(3):162176.
[158] Li J, Blumenfeld DE, Marin SP (2007) Manufacturing System Design to
Improve Quality Buy Rate: An Automotive Paint Shop Application Study.
IEEE Transactions on Automation Science and Engineering 4(1):7579.
[159] Li J, Blumenfeld DE, Huang N, Alden JM (2009) Throughput Analysis of
Production Systems: Recent Advances and Future Topics. International Journal of Production Research 47(14):38233851.

[160] Li J, Blumenfeld E (2006) Quantitative Analysis of a Transfer Production Line

with Andon. IIE Transactions 38(10):837846.
[161] Li J, Huang N (2007) Quality Evaluation in Flexible Manufacturing Systems: A
Markovian Approach. Mathematical Problems in Engineering 2007.
[162] Li J, Jin J, Shi J (2008) Causation-Based t2 Decomposition for Multivariate
Process Monitoring and Diagnosis. Journal of Quality Technology 40(1):46.
[163] Li L, Chang Q, Ni J, Xiao G, Biller S (2007) Bottleneck Detection of Manufacturing Systems Using Data Driven Method. Assembly and Manufacturing, ISAM 07
IEEE International Symposium, 7681.
[164] Li L, Wareld JN (2011) Perspectives on Quality Coordination and Assurance in
Global Supply Chains. International Journal of Production Research 49(1):14.
[165] Li W (2005) Comparative Study of Push and Pull Systems Considering Quality
Performance in a Cell-Based Job Shop Environment. International Journal of
Production Research 43(23):49995017.
[166] Liana SY, Hecker RL, Landers RG (2004) Machining Process Monitoring and
Control: The State-of-the-Art. Journal of Manufacturing Science and Engineering 126(2):297310.
[167] Liberopoulos G (2013) Production Release Control: Paced, WIP-Based, or
Demand-Driven? Revisiting the Push/Pull and Make-to-Order/Make-toStock Distinctions.Smith JM, Tan B, (Eds.) Handbook of Stochastic Models
and Analysis of Manufacturing System Operations International Series in Operations Research and Management Science, vol. 192. Springer, New York, NY

211247.
[168] Liberopoulos G, Kozanidis G, Tsarouhas P (2007) Performance Evaluation of
an Automatic Transfer Line with WIP Scrapping During Long Failures.
Manufacturing and Service Operations 9(1):6283.
[169] Liberopoulos G, Tsarouhas PH (2002) Systems Analysis Speeds Up Chipitas
Food Processing Line. Interfaces 32(3):6276.
[170] Linderman K, McKone-Sweet KE, Anderson JC (2005) An Integrated Systems
Approach to Process Control and Maintenance. European Journal of Operational Research 164:324340.
[171] Liu J (2010) Variation Reduction for Multistage Manufacturing Processes: A Comparison Survey of Statistical-Process-Control vs Stream-ofVariation Methodologies. Quality and Reliability Engineering International
26(7):645661.
[172] Liu J, Shi J, Hu SJ (2009) Quality-Assured Setup Planning Based on the Streamof-Variation Model for Multi-Stage Machining Processes. IIE Transactions
41(4):323334.
[173] Llinas J, Hall DL (1997) An Introduction to Multi-sensor Data Fusion. Proceedings of IEEE 623.
[174] Lochte C, Kayasa J, Herrmann C, Raatz A (2012) Methods for Implementing
Compensation Strategies in Micro Production Systems Supported by a Simulation Approach. Precision Assembly Technologies and Systems 371:118125.
[175] Loose J, Zhou S, Ceglarek D (2007) Kinematic Analysis of Dimensional
Variation Propagation for Multistage Machining Processes with General
Fixture Layouts. IEEE Transactions on Automation Science and Engineering
4(2):141152.
[176] Ma YHK, Koren Y (2004) Operation of Manufacturing Systems with Work-inprocess Inventory and Production Control. CIRP Annals Manufacturing
Technology 53(1):361365.
[177] Makris S, Pintzos G, Rentzos L, Chryssolouris G (2013) Assembly Support
Using AR Technology Based on Automatic Sequence Generation. CIRP Annals
Manufacturing Technology 62(1):913.
[178] Mandroli SS, Shrivastava AK, Ding Y (2006) A Survey of Inspection Strategy
and Sensor Distribution Studies in Discrete-Part Manufacturing Processes. IIE
Transactions 38(4):309328.
[179] Mannar K, Ceglarek D, Niu F, Abifaraj B (2006) Fault Region Localization:
Product and Process Improvement Based on Field Performance and
Manufacturing Measurements. IEEE Transactions on Automation Science and
Engineering 3(4):423439.
[180] Maropoulos PG, Ceglarek D (2010) Design Verication and Validation in
Product Lifecycle. CIRP Annals Manufacturing Technology 59(2):740759.
[181] McCall JJ (1963) Operating Characteristics of Opportunistic Replacement and
Inspection Policies. Management Science 10(1):8597.
[182] McDonald T, Ellis KP, Van Aken EM, Patrick Koelling C (2009) Development
and Application of a Worker Assignment Model to Evaluate a Lean
Manufacturing Cell. International Journal of Production Research 47(9):
24272447.
[183] McKinsey&Company (2013) Manufacturing the Future: The Next Era of Global
Growth and Innovation.
[184] Mease DA, Sudjinato V, Nair N (2004) Selective Assembly in Manufacturing: Statistical Issues and Optimal Binning Strategies. Technometrics
46(2):165175.
[185] Meerkov SM, Zhang L (2010) Product Quality Inspection in Bernoulli Lines:
Analysis, Bottlenecks, and Design. International Journal of Production Research
48(16):47454766.
[186] Meerkov SM, Zhang L (2011) Bernoulli Production Lines with QualityQuantity Coupling Machines: Monotonicity Properties and Bottlenecks. Annals of
Operations Research 182(1):119131.
[187] Meier H, Roy R, Seliger G (2010) Industrial Product-Service Systems-IPS2.
CIRP Annals Manufacturing Technology 59(2):607627.
[188] Mhada FZ, Malhame RP, Pellerin R (2013) Joint Assignment of Buffer Sizes
and Inspection Points in Unreliable Transfer Lines with Scrapping of Defective Parts. Production & Manufacturing Research 1(1):79101.
[189] Miltenburg J (2001) U-Shaped Production Lines: A Review of Theory and
Practice. International Journal of Production Economics 70(3):201214.
[190] Miropolsky A, Fischer A (2006) A Uniform Approach for Utilizing Synergy
between Inspection Technologies and Computational Methods. CIRP Annals
Manufacturing Technology 55(1):123126.
[191] Monostori L, Vancza J, Kumara S (2006) Agent-Based Systems for
Manufacturing. CIRP Annals Manufacturing Technology 55(2):97720.

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

[192] Montgomery DC, Woodall WH (1997) A Discussion on Statistically-Based
Process Monitoring and Control. Journal of Quality Technology 29(2):121162.
[193] Mori M, Fujishima M, Komatsu M, Zhao B, Liu Y (2008) Development of
Remote Monitoring and Maintenance System for Machine Tools. CIRP Annals
Manufacturing Technology 57(1):433436.
[194] Moroni G, Pacella M (2008) An Approach Based on Process Signature Modeling for Roundness Evaluation of Manufactured Items. Journal of Computing
and Information Science in Engineering 8(2).
[195] Mourtzis D, Doukas M (2013) Decentralized Manufacturing Systems Review:
Challenges and Outlook. Robust Manufacturing Control, Springer: 355369.
[196] Mourtzis D, Doukas M, Psarommatis F (2012) A Multi-Criteria Evaluation of
Centralized and Decentralized Production Networks in a Highly CustomerDriven Environment. CIRP Annals Manufacturing Technology 61(1):427430.
[197] Mourtzis D, Papakostas N, Makris S, Xanthakis V, Chryssolouris G (2008)
Supply Chain Modeling and Control for Producing Highly Customized Products. CIRP Annals Manufacturing Technology 57(1):451454.
[198] Muller A, Crespo Marquez A, Iung B (2008) On the Concept of e-Maintenance,
Review and Current Research. Reliability Engineering and System Safety
93(8):11651187.
[199] Muller A, Suhner MC, Iung B (2008) Formalisation of a New Prognosis Model
for Supporting Proactive Maintenance Implementation on Industrial System.
Reliability Engineering & System Safety 93(2):234253.
[200] MuProD, Innovative proactive Quality Control system for in-process multistage defect reduction, European Union FP7. Grant agreement n8 285075.
[201] Nada OA, ElMaraghy HA, ElMaraghy WH (2006) Quality Prediction in
Manufacturing System Design. Journal of Manufacturing systems 25(3):153171.
[202] Nee AYC, Ong SK, Chryssolouris G, Mourtzis D (2012) Augmented Reality
Applications in Design and Manufacturing. CIRP Annals Manufacturing
Technology 61(2):657679.
[203] Newman ST, Nassehi A (2009) Machine Tool Capability Prole for Intelligent
Process Planning. CIRP Annals Manufacturing Technology 58:421424.
[204] Ni J, Jin X, Koren Y (2012) Decision Support Systems for Effective Maintenance
Operations. CIRP Annals Manufacturing Technology 61(1):411414.
[205] Nicolai R, Dekker R (2008) Optimal Maintenance of Multi-component Systems:
A Review. Complex System Maintenance Handbook, Springer: 263286.
[206] Nikon. On-line Documentation, www.nikonmetrology.com.
[207] NIST (2008) NIST IEEE-P1451 Draft Standard Home Page. http://ieee1451.nist.
gov/accessed (accessed 1.10.09).
[208] Nonaka Y, Erdo s G, Kis T, Kovacs A, Monostori L, Nakano T, Vancza J (2013)
Generating Alternative Process Plans for Complex Parts. CIRP Annals
Manufacturing Technology 62(1):453458.
[209] Nonaka Y, Erdo s G, Kis T, Nakano T, Vancza J (2012) Scheduling with
Alternative Routings in CNC Workshops. CIRP Annals Manufacturing Technology 61(1):449454.
[210] Nonaka Y, Suginishi Y, Lengyel A, Ono M, Sugimoto K (2008) TSUNAMI Effect
Prediction Methodology for Critical Resource Analysis. Manufacturing Systems and Technologies for the New Frontier, Springer, London337340.
[211] Nourelfath ME, Chatelet (2012) Integrating Production, Inventory and Maintenance Planning for a Parallel System with Dependent Components. Reliability Engineering & System Safety 101:5966.
[212] Ong SK, Zhu J (2013) A Novel Maintenance System for Equipment Serviceability Improvement. CIRP Annals Manufacturing Technology 39.
[213] Osogami T, Harchol Balter M (2006) Closed Form Solutions for Mapping
General Distributions to Quasi-minimal PH Distributions. Performance Evaluation 62:524552.
[214] Owen H, Blumenfeld E (2008) Effects of Operating Speed on Production Quality
and Throughput. International Journal of Production Research 46(24):70397056.
[215] Panagiotidou S, Tagaras G (2012) Optimal Integrated Process Control and
Maintenance under General Deterioration. Reliability Engineering & System
Safety 104:5870.
[216] Pandey D, Kulkarni MS, Vrat P (2011) A Methodology for Joint Optimization
for Maintenance Planning, Process Quality and Production Scheduling. Computers & Industrial Engineering 61(4):10981106.
[217] Pellegrinelli S, Borgia S, Tolio T (2010) Step-NC Compliant Approach for
Workpiece Setup Planning Problem on Transfer Line. Proceedings of 43rd CIRP
Conference on Manufacturing Systems, 433440IBN: 978-3-7083-0686-5.
[218] Pellegrinelli S, Tolio T (2013) Pallet Operation Sequencing Based on Network Part
Program Logic. Robotics and Computer-Integrated Manufacturing 29(5):322345.
[219] Podshivalov L, Fischer A, Bar-Yoseph PZ (2010) Multiresolution 2D Geometric
Meshing for Multiscale Finite Element Analysis of Bone Micro-Structures.
Virtual and Physical Prototyping 5(1):3343.
[220] Prakash. Izquierdo LE, Ceglarek D (2009) Functional Process Adjustments to
Reduce No-Fault-Found Product Failures in Service Caused by In-Tolerance
Faults. CIRP Annals Manufacturing Technology 58(1):3740.
[221] Prakash PKS, Ceglarek D (2013) Multi-Step Functional Process Adjustments
to Reduce No-Fault-Found Product Failures in Service Caused by In-tolerance
Faults. Procedia CIRP 11:3843.
[222] Prandtstetter M, Raidl GR (2008) An Integer Linear Programming Approach
and a Hybrid Variable Neighborhood Search for the Car Sequencing Problem.
European Journal of Operational Research 191(3):10041022.
[223] Putz M, Langer T (2012) Determination of Extended Availability and Productivity for Assembly Systems Using Existing Database. CIRP Annals
Manufacturing Technology 61(1):1720.
[224] Radhoui M, Rezg N, Chelbi A (2009) Integrated Model of Preventive Maintenance, Quality Control and Buffer Sizing for Unreliable and Imperfect Production Systems. International Journal of Production Research 47(2):389402.
[225] Raz T (1986) A Survey of Model for Allocation Inspection Effort in Multistage
Production System. Journal of Quality Technology 18(4):239247.
[226] Reinhart G, Tekouo W (2009) Automatic Programming of Robot-Mounted 3D
Optical Scanning Devices to Easily Measure Parts in High-Variant Assembly.
CIRP Annals Manufacturing Technology 58(1):2529.

795

[227] Renishaw (2008) Probing Systems for CNC Machine Tools, Renishaw Plc: 172.
[228] RLW Navigator, FP7 2011, European Project Remote Laser Welding System
Navigator for Eco & Resilient Automotive Factories. Grant Agreement 285051.
[229] RobustPlaNet, FP7 2013 European Project RobustPlaNet, Shock-Robust Design of Plants and their Supply Chain Networks. Grant Agreement 609087.
[230] Repenning NP, Sterman JD (2002) Capability Traps and Self-Conrming
Attribution Errors in the Dynamics of Process Improvement. Administrative
Science Quarterly 47(2):265295.
[231] Robson K, MacIntyre J, Trimble R (2013) Measuring the status and
alignment of maintenance and manufacturing strategiesthe development of a new model and diagnostic tool. Journal of Quality in Maintenance
Engineering 381397.
[232] Rosenzweig ED, Easton GS (2010) Tradeoffs in Manufacturing? A Meta-Analysis and
Critique of the Literature. Production and Operations Management 19(2):127141.
[233] Rotondo A, Young P, Geraghty J (2012) Quality Risk Prediction at a NonSampling Station Machine in a Multi-Product, Multi-Stage, Parallel Processing Manufacturing System Subjected to Sequence Disorder and Multiple
Stream Effects. Annals of Operations Research 209(1):255277.
[234] Ruifeng C, Subramaniam V (2011) Increasing Production Rate in Kanban
Controlled Assembly Lines Through Preventive Maintenance. International
Journal of Production Research 118.
[235] Samak D, Fischer A, Rittel D (2007) 3D Reconstruction and Visualization of
Microstructure Surfaces from 2D Images. CIRP Annals Manufacturing Technology 56(1):149152.
[236] Sarker BR, Jamal AMM, Mondal S (2008) Optimal Batch Sizing in a MultiStage Production System with Rework Consideration. European Journal of
Operational Research 184(3):915929.
[237] Satow T, Osaki S (2003) Opportunity-Based Age Replacement with Different
Intensity Rates. Mathematical and Computer Modelling 38(1113):14191426.
[238] Schick IC, Gershwin SB, Kim J (2005) Quality/Quantity Modeling and Analysis
of Production Lines Subject to Uncertainty. Final Report Presented to General
Motors Research and Development Center.
[239] Schick. Irvin C, Gershwin. Stanley B (2007) A Taxonomy of Quality/Quantity
Issues in Manufacturing Systems. Proceedings of the 6th Conference on the
Analysis of Manufacturing Systems.
[240] Schmitt R, Isermann M, Wagels C (2001) Integrative Self-Optimizing Process
Chains. in Brecher C, (Ed.) Integrative Production Technology for High-Wage
Countries, Springer, Berlin.
[241] Schmitt R, Beaujean P (2009) The Quality Backward ChainThe Adaptive
Controller of Entrepreneurial Quality. Proceedings of the 6th CIRP-Sponsored
International Conference on Digital Enterprise Technology, 11331143.
[242] Schmitt R, Falk B, Quattelbaum B (2009) Perceived Quality New Information
Data for Production Specications. CAT 2009 11th CIRP International Conference on Computer Aided Tolerancing: Geometric Variations within Product
Life-Cycle Management.
[243] Schmitt R, Falk B, Quattelbaum B, Knispel J (2013) Tolerancing Subjective and
Uncertain Customer Requirements Regarding Perceived Product Quality.
Journal of Engineering Manufacture 227(5):702708.
[244] Schmitt R, Isermann M, Niggemann C, Laass M, Matuschek N (2010) Cognition-Based Self-Optimisation of an Automotive Rear-Axle-Drive Production
Process. Journal of Machine Engineering 10(3):6877.
[245] Schmitt R, Kristes D, Beaujean P (2008) Entrepreneurial Quality Management
A New Denition of Quality. 2008 IEEE International Engineering Management Conference, 275280.
[246] Schmitt R, Basse I (2011) A Holistic Approach to Decision Theory-Based
Inspection Planning. Procceedings 14th QMOD Conference on Quality and
Service Sciences 2011, San Sebastian, Spain, August 2931, 214224.
[247] Schmitt R, Monostori L, Glockner H, Viharos Z (2012) Design and Assessment
of Quality Control Loops for Stable Business Processes. CIRP Annals
Manufacturing Technology 61(1):439444.
[248] Schmitt R, Niggemann C, Laass M (2012) Cognitive Failure Cluster enhancing the
Efciency and the Precision of the Self-Optimizing Process Modell for Bevel Gear
Contact Patterns. Journal of Machine Engineering 12(1):5565.
[249] Schmitt R (2012) Self-Optimizing production systems. in Brecher C, (Ed.) Integrative Production Technology for High-Wage Countries, Springer, pp. 698891.
[250] Schmitt R, Janssen C, Basse I, Schmitt S (2013) Model for a Decision Theory
Based Inspection Planning. in Zaeh M, (Ed.) Enabling Manufacturing Competitiveness and Economic Sustainability Proceedings of the 5th International
Conference on Changeable, Agile, Recongurable and Virtual Production
CARV2013), Springer, Berlin383388.
[251] Schmitt R, Wagels C (2012) Benchmarking of Methods and Instruments for
Self-Optimization in Future Production Systems. Procedia CIRP of 45th CIRP
Conference on Manufacturing Systems, 161166.
[252] Schmitt R, Wagels C, Isermann M (2008) Cognitive Tolerance Matching.
Proceedings of the 6th CIRP International Conference on Intelligent Computation
in Manufacturing Engineering:, 289293.
[253] Scholz-Reiter B, Freitag M, Schmieder A, Pikovsky A, Katzorke I (2006) in
Radons G, Neugebaue R, (Eds.) Nonlinear Dynamics of Production Systems, Wiley.
[254] Scholz-Reiter B, Gorges M, Philipp T (2009) Autonomously Controlled Production Systems Inuence of Autonomous Control Level on Logistic Performance. CIRP Annals Manufacturing Technology 58(1):395398.
[255] Scholz-Reiter B, Rekersbrink H, Gorges M (2010) Dynamic Flexible Flow Shop
Problems Scheduling Heuristics vs. Autonomous Control. CIRP Annals
Manufacturing Technology 59(1):465468.
[256] Scholz-Reiter B, Freitag M (2007) Autonomous Processes in Assembly Systems. CIRP Annals Manufacturing Technology 56(2):712729.
[257] Seleim A, ElMaraghy HA (2014) Transient Analysis of a Re-entrant Manufacturing
System. Enabling Manufacturing Competitiveness and Economic Sustainability, 9195.
[258] Seliger G, Consiglio S, Zettl M (2004) Selling Use Instead of Selling Products:
Technological and Educational Enablers for Business in Ecological Product
Life Cycles. Production Engineering and Computers 6(7):1721.

796

M. Colledani et al. / CIRP Annals - Manufacturing Technology 63 (2014) 773796

[259] Shanoun M, Bassetto S, Bastoini S, Vialletelle P (2011) Optimisation of the Process

Control in a Semiconductor Company: Model and Case Study of Defectivity
Sampling. International Journal of Production Research 49(13):38733890.
[260] Shi C, Gershwin S (2012) Part Waiting Time Distribution in a Two-Machine
Line. Information Control Problems in Manufacturing 14(1):457462.
[261] Shi J (2006) Stream of Variation Modeling and Analysis for Multistage
Manufacturing Processes, CRC Press, Taylor & Francis.
[262] Shi J, Zhou S (2009) Quality Control and Improvement for Multistage Systems: A Survey. IIE Transactions 41(9):744753.
[263] Shmukler A, Fischer A (2009) Verication of 3D Freeform Parts by Registration of Multi-Scale Shape Descriptors. International Journal of Advanced
Manufacturing Technology 49:10931106.
[264] Shu MH, Hsu BM, Kapur KC (2010) Dynamic Performance Measures for Tools
with Multi-State Wear Processes and their Applications for Tool Design and
Selection. International Journal of Production Research 48(16):47254744.
[265] Siener M, Aurich JC (2011) Quality Oriented Maintenance Scheduling. CIRP
Journal of Manufacturing Science and Technology 4(1):1523.
[266] Sila I, Ebrahimpour M, Birkholz C (2006) Quality in Supply Chains: An Empirical
Analysis. Supply Chain Management: An International Journal 11(6):491502.
[267] Sloan TW, Shanthikumar GJ (2000) Combined Production and Maintenance
Scheduling for a Multiple-Product, Single-Machine Production System. Production and Operations Management 9(4):379399.
[268] Sloan TW, Shanthikumar GJ (2002) Using In-Line Equipment Condition and
Yield Information for Maintenance Scheduling and Dispatching in Semiconductor Wafer Fabs. IIE Transactions 34:191209.
[269] Stecke KE, Aronson JE (1985) Review of Operator & Sol: Machine Interference
Models. International Journal of Production Research 23(1):129151.
[270] Sterman JD (2000) Business Dynamics, Systems Thinking and Modeling for a
Complex World, Irwing/McGraw Hill.
[271] Su Q, Liu L, Whitney DE (2010) A Systematic Study of the Prediction Model for
Operator-Induced Assembly Defects Based on Assembly Complexity Factors.
IEEE Transactions on Systems Man and Cybernetics Part A: Systems and Humans
40(1):107120.
[272] Subbaiah KV, Rao SVUM, Rao KS (2011) An Inventory Model for Perishable
Items with Alternating Rate of Production. International Journal of Advanced
Operations Management 3(1):6687.
[273] Subramaniam SK, Husin SH, Singh R, Hamidon A (2009) Production Monitoring System for Monitoring the Industrial Shop Floor Performance. International Journal of Systems Applications Engineering Development 3(1):2835.
[274] Takata S, Kirnura F, Van Houten FJAM, Westkamper E, Shpitalni M, Ceglarek
D, Lee J (2004) Maintenance: Changing Role in Life Cycle Management. CIRP
Annals Manufacturing Technology 53(2):643655.
[275] Tan B (2013) Modeling and Analysis of Output Variability in Discrete Material
Flow Production Systems. Handbook of Stochastic Models and Analysis of
Manufacturing System Operations, Springer, New York287311.
[276] Tansky A, Fischer B, Colosimo M, Ben Shabat Y (2013) Multi-Sensor MultiResolution Data Fusion Modeling. CIRP Design International Conference on
Smart Product Engineering, Bochum, Germany.
[277] Teti R, Jemielniak K, ODonnell G, Dornfeld D (2010) Advanced Monitoring of
Machining Operations. CIRP Annals Manufacturing Technology 59(2):717739.
[278] Tolio T (2009) Design of Flexible Manufacturing Systems, Springer.
[279] Tolio T (2013) Interactions Among Production Logistics, Quality and Maintenance in Manufacturing Systems. Keynote Speech at the SMMSO 2013
Conference on Stochastic Models for Manufacturing and Service Operations.
[280] Tolio T, Ceglarek D, Elmaraghy HA, Fischer A, Hu SJ, Laperriere L, Newman ST,
Vancza J (2010) SPECIES-Co-Evolution of Products, Processes and Production
Systems. CIRP Annals Manufacturing Technology 59(2):672693.
[281] Toshniwal V, Dufe N, Jagalski T, Rekersbrink H, Scholz-Reiter B (2011)
Assessment of Fidelity of Control-Theoretic Models of WIP Regulation in
Networks of Autonomous Work Systems. CIRP Annals Manufacturing Technology 60(1):485488.
[282] Tseng MM, Jiao RJ, Wang C (2010) Design for Mass Personalization. CIRP
Annals Manufacturing Technology 59(1):175178.
[283] Tsung F, Li Y, Jin M (2008) Statistical Process Control for Multistage
Manufacturing and Service Operations: A Review and Some Extensions.
International Journal of Services Operations and Informatics 3(2):191204.
[284] Tubilla F, Gershwin S (2009) The Hedging Zone Policy for Real-Time Scheduling of Sequence-Dependent Setups. Information Control Problems in
Manufacturing 13(1):14871491.
[285] Urban TL (1998) Analysis of Production Systems When Run Length Inuences
Product Quality. International Journal of Production Research 36(11):30853094.
[286] Van der Duyn Schouten FA, Vanneste SG (1995) Maintenance Optimization of
a Production System with Buffer Capacity. European Journal of Operational
Research 85:323338.
[287] Van Volsem S, Dullaert W, Van Landeghem H (2007) An Evolutionary
Algorithm and Discrete Event Simulation for Optimizing Inspection Strategies for Multi-Stage Processes. European Journal of Operational Research
179(3):621633.
[288] Vancza J, Egri P (2006) Coordinating Supply Networks in Customized Mass
Production A Contract-Based Approach. CIRP Annals Manufacturing Technology 55(1):489492.
[289] Vancza J, Egri P, Monostori L (2008) A Coordination Mechanism for Rolling
Horizon Planning in Supply Networks. CIRP Annals Manufacturing Technology 57(1):455458.
[290] Vancza J, Monostori L, Lutters D, Kumara SR, Tseng M, Valckenaers P, Van
Brussel H (2011) Cooperative and Responsive Manufacturing Enterprises.
CIRP Annals Manufacturing Technology 60(2):797820.
[291] Venkatasubramanian V (2005) Prognostic and Diagnostic Monitoring of
Complex Systems for Product Lifecycle Management: Challenges and Opportunities. Computers & Chemical Engineering 29(6):12531263.

[292] Vijayaraghavan A, Sobel W, Fox A, Dornfeld D, Warndorf P (2008) Improving

Machine Tool Interoperability Using Standardized Interface Protocols: MTConnect. International Symposium on Flexible Automation, ASME, pp. 2326.
[293] Waiyagan K, Bohez EL (2009) Intelligent Feature Based Process Planning for
Five-Axis Mill-Turn Parts. Computers in Industry 60(5):296316.
[294] Wang H (2002) A Survey of Maintenance Policies of Deteriorating Systems.
European Journal of Operational Research 132:469489.
[295] Wang J, Hu Y, Li J (2010) Transient Analysis to Design Buffer Capacity in Dairy
Filling and Packing Production Lines. Journal of Food Engineering 98(1):112.
[296] Wang J, Li J, Arinez J, Biller S (2013) Quality Bottleneck Transitions in Flexible
Manufacturing Systems with Batch Productions. IIE Transactions 45(2):190205.
[297] Wang J, Li J, Arinez J, Biller S (2010) Product Sequencing with Respect to
Quality in Flexible Manufacturing Systems with Batch Operations. Automation Science and Engineering IEEE Transactions on 7(4):776790.
[298] Wang J, Li J, Arinez J, Biller S, Huang N (2010) Quality Analysis in Flexible
Manufacturing Systems with Batch Productions: Performance Evaluation
and Non-Monotonic Properties. Automation Science and Engineering IEEE
Transactions on 7(3):671676.
[299] Wang J, Li J, Arinez J, Biller S (2012) Indicators for Quality Improvability and
Bottleneck Sequence in Flexible Manufacturing Systems with Batch Productions. International Journal of Production Research 50(22):63886402.
[300] Wang KH, Chen WL, Yang DY (2009) Optimal Management of the Machine
Repair Problem with Working Vacation: Newtons Method. Journal of Computational and Applied Mathematics 233(2):449458.
[301] Wang K, Han K (2013) A Batch-Based Run-To-Run Process Control Scheme for
Semiconductor Manufacturing. IIE Transactions 45(6):658669.
[302] Wang YM, Nagarkar SR (1999) Locator and Sensor Placement for Automated
Coordinate Checking Fixtures. Journal of Manufacturing Science and Engineering 121(4):709719.
[303] Weber A (2004) Is Flexibility a Myth? Assembly 47(5):5059.
[304] Weckenmann A, Jiang X, Sommer KD, Neuschaefer-Rube U, Seewig J, Shaw L,
Estler T (2009) Multisensor Data Fusion in Dimensional Metrology. CIRP
Annals Manufacturing Technology 58(2):701721.
[305] Wegener K, Hoffmeister HW, Karpuschewski B, Kuster F, Hahmann WC,
Rabiey M (2011) Conditioning and Monitoring of Grinding Wheels. CIRP
Annals Manufacturing Technology 60(2):757777.
[306] Wiendahl HP, Andersson P, Berggre C, Nehler C (2004) Manufacturing Firms
and Integrated Solutions: Characteristics and Implications. European Journal
of Innovation Management 7(1):218228.
[307] Wiendahl HP, Kuprat T (1991) Logistic Analysis of Production Processes by
Operating Curves. CIRP Annals Manufacturing Technology 40(1):475478.
[308] Wiendahl HP, von Cieminski G, Begemann C, Nickel R (2005) Human Factors
in Production Planning and Control. Integrating Human Aspects in Production
Management, vol. 160. 113125.
[309] Wiendahl H-P, ElMaraghy H, Nyhuis P, Zah M, Wiendahl H-H, Dufe N, Brieke
M (2007) Changeable Manufacturing Classication, Design and Operation.
CIRP Annals Manufacturing Technology 56(2):783809.
[310] Wright CM, Mehrez A (1998) An Overview of Representative Research of the
Relationships Between Quality and Inventory. Omega 26(1):2947.
[311] Wu Z, Shamsuzzaman M, Wang Q (2007) The Cost Minimization and Manpower Deployment to SPC in a Multistage Manufacturing System. International Journal of Production Economics 106(1):275287.
[312] Xiang L, Tsung F (2008) Statistical Monitoring of Multistage Processes Based
on Engineering Models. IIE Transactions 40(10):957970.
[313] Xu X, Wang L, Newman ST (2011) Computer-Aided Process Planning A
Critical Review of Recent Developments and Future Trends. International
Journal of Computer Integrated Manufacturing 24(1):131.
[314] Yan CB, Hassoun M, Meerkov SM (2012) Equilibria, Stability, and Transients
in Re-entrant Lines Under FBFS and LBFS Dispatch and Constant Release.
Semiconductor Manufacturing IEEE Transactions on 25(2):211229.
[315] Yang L, Lee J (2012) Bayesian Belief Network-Based Approach for Diagnostics
and Prognostics of Semiconductor Manufacturing Systems. Robotics and
Computer-Integrated Manufacturing 28(1):6674.
[316] Yang ZM, Djurdjanovic D, Ni J (2008) Maintenance Scheduling in Manufacturing Systems Based on Predicted Machine Degradation. Journal of Intelligent
Manufacturing 19(1):8798.
[317] Yang Z, Djurdjanovic D, Ni J (2007) Maintenance Scheduling for a
Manufacturing System of Machines with Adjustable Throughput. IIE Transactions 39(12):11111125.
[318] Yano CA, Lee HL (1995) Lot Sizing and Random Yields: A Review. Operations
Research 43:311334.
[319] Yao X, Xie X, Fu MC, Marcus SI (2005) Optimal Joint Preventive Maintenance
and Production Policies. Naval Research Logistics 52(7):668681.
[320] Yeh RH (1997) State-Age-Dependent Maintenance Policies for Deteriorating
Systems with Erlang Sojourn Time Distributions. Reliability Engineering System Safety 58:5560.
[321] Yeung TG, Cassady CR, Schneider K (2008) Simultaneous Optimization of
Control Chart and Age-Based Preventive Maintenance Policies Under an
Economic Objective. IIE Transactions 40:147159.
[322] Zaideman O, Fischer A (2010) Geometric Bone Modeling: From Macro to
Micro-Structures. Journal of Computer Science and Technology 25(3):614622.
[323] Zero WIN Towards Zero-Waste in Industrial Networks, European Community, FP7/2007-2013. Grant agreement n8 226752.
[324] Zhou J, Djurdanovic D, Ivy J, Ni J (2007) Integrated Reconguration and Age-Based
Preventive Maintenance Decision Making. IIE Transactions 39(12):10851102.
[325] Zhu X, Hu SJ, Koren Y, Marin SP (2008) Modeling of Manufacturing Complexity in Mixed-Model Assembly Lines. Journal of Manufacturing Science and
Engineering 130(5):51013.
[326] Zou C, Tsung F, Liu Y (2008) A Change Point Approach for Phase I Analysis in
Multistage Processes. Technometrics 50(3):344356.